Why doesn't HS consider deck composition?

There is only a benefit to having the matchmaker consider a deck’s power level, ensuring matches against decks of similar strength.

Hearthstone’s failure to implement this is detrimental to the casual experience, which is essential for a game like Hearthstone that lacks the complexity of other games.

Additionally, the bot problem stems from the matchmaker’s disregard for deck power levels. If it did consider them, players wouldn’t be matched against decks hard running Murloc Tidehunter and such, includes which bots commonly use because of their disregard for competitive play but only for accumulation of experience. Like that, bots would only be matched against other bots.

Deck composition should be the primary criterion for matchmaking. Also, Bonus Star system should be removed, while only queueing players in their representative rank. Good players would climb out of it while proving of their skill by beating decks of the same power level. Therefore, Legend would be filled with genuine good players, and not tier 0 deck abusers. They would prove themselves on a level playing field.

Also, it would make queueing less triggering.
If the matchmaker reviewed your match history and noticed a sub-50% win rate against, for instance, Brann Warrior, it would pair you against a different deck to help bring your win rate closer to 50%. This adjustment would create a more enjoyable experience for the player which hates playing against X deck, since they would have agency over facing it less.

2 Likes

So you want the matchmaking to be rigged.

7 Likes

Because it doesn’t have to, your results are enough.

If a thousand people all have a 50% win rate against each other, and you start facing them and have a 50% win rate against them too, then you’re being matched appropriately no matter what decks are being played.

If someone in that thousand makes a better deck and starts winning more, then they stop playing against that thousand and start playing against a different thousand set of higher-rated players. It wasn’t the deck that moved them up, it was the wins.

Results are all that is necessary, as decades of matching systems in numerous games have shown.

3 Likes

Results are crucial, but they shouldn’t overshadow the potential benefits of incorporating deck composition into matchmaking to foster a more enjoyable gaming experience.

In my scenario, there is both results, and that.

Further, another result of the MMR system, is the bot problem, which isn’t exactly an intended result.

1 Like

I agree, the matchmaking seems off. I get paired against decks that will defeat me easily. It almost feels rigged for the opponents to win and for me to lose. I barely experience a winning spree and when I do win, it feels good but then I’m left losing multiple games in a row. It doesn’t feel like a 50/50 percent win chance. It’s feels like some people are more favored to win and I’m not sure why my hunter deck being as good as it is said to be, doesn’t feel like it is. Feels like this game favors druids and priest more than they should. Maybe the company needs to investigate why these classes have cards that are stronger or create more of an unfair advantage. It feels like someone in the company purposely makes outrageous cards that shouldn’t exist. I’m on the verge of uninstalling this game but it’s not like company cares about 1 person like me as long as they’re fooling everyone else and taking their money. Plus these people that comment on here usually are developers or people who just don’t experience what we experience. A lot of people that comment on here are one sided and most of the time, don’t understand what other people go through.

More enjoyable how, and for whom?

How do you know when you succeed? By whose standards? How often do you need to mix things up? Should you mix things up? Can you do this all and still make fair matches?

You are drastically underestimating the complexity involved. The implementation, effects, results to players, all are entirely subjective and wholly unmeasurable.

You will almost certainly make things worse, because you can’t define what it means to be better.

4 Likes

Would it be better if both players were able to see not only the opponents rank in legend, but for everyone to know each other’s rank, star bonus, and the elusive unknown mmr?
That along with an understanding of how the matches are calculated as clearly explained by several posters on this forum.
I just wondered what the playerbase thinks about this suggestion.

1 Like

Everyone, you are matched on a level playing field.

There’s no need to argue that winning and achieving greatness is more satisfying when
with a handicap, it’s just a video game. If Blizzard has the ability, they should ensure that everyone plays on an equal footing for a better experience and to confirm that any losses are indeed their responsibility.

Their income will be the card acquisition, but the playing field should be made to ensure balance. If I queue a “fun” deck right now, with my MMR I will just be facing the same opponents I would with the actually good deck that brought me to this MMR. That’s not enjoyable, I can tell you that. Neither is it “fair.”

I prefer they take their “experimentation” to stuff like this, rather than failed projects like Twist, Mercenaries and what not.

I am not underestimating it at all. You are assuming.

Better is to ensure you aren’t being “cheated.”
Grand scheme of “better” in a lax game like HS.

MMR system won’t provide that, since you can queue a Tier 1 deck, and a Tier 4 deck and you’ll face the same opponents.

Doesn’t ensure which player is better, and neither will my system since variance remains. It will at least be closer to it, though.

The problem is that if the system takes your deck into account, it’s only a matter of time until people figure out the rules to it in order to exploit it and make the experience worse for everyone.

It’s nearly impossible to set up a MMR system that also takes your deck into account that doesn’t have really awkward quirks that kind of ruin the whole thing.

1 Like

Exploit in what way?

If someone is able to create/find a deck that is seemingly “weak” by the matchmaker’s standards and pushes with it to a higher realm, I wouldn’t say it’s exploiting the system.

Remember the matchmaker will make his analysis based on general win rate, so the decks will still match in power level. There is just room for skill expression for that deck.

It’s rewarding player ingenuity.

Also, the seemingly “weak” deck thing is happening today as well.
Decks that seem weak to the general populous become way stronger in the hands of an experienced player.

My idea is that MMR as a whole should be removed.

Player should be matched solely based on deck power, and ranking.

Good players will progress, and the losing players will have nobody to blame but themselves, since the system ensures fair matchmaking. As time goes on and the good players progress the ranks, the losing players will start facing each other so it wouldn’t even make for a bad play experience for the losing side.

There will be the off-chance of a good player facing a bad player, but as long as their decks are made to provide a fair matchup, the bad player shouldn’t even complain. And still, that would be the off-chance. Good players won’t be in these ranks for long since their skill expression will beat your average X rank player, and make them go beyond that rank.

You find the strongest deck for its cutoff and use it to reign over everything under it. Or you find a way to build your deck that dodges your bad matchup.

Or you end up with weird pockets with only mirror matches.

The game starts to become how you build your deck around the matchmaking system more than building a good deck and piloting it well in all circumstances.

Also, if you removed mmr entirely, id suddenly have like 80-90% win rates when playing seriously, as I’d have a massive pool of players that make terrible decisions to farm.

Your build won’t shake the matchmaker.
For Blizzard’s sake to create an advanced machine like so, your deck’s game plan will be considered, not each individual card. They wouldn’t create something that advanced.

It will amount to Rainbow Mage versus Brann Warrior.

Now you taking advantage of that “loop” to slip cards that are good against your bad matchup without the matchmaker being able to change its decision based on that, isn’t a bad thing.

The range is big enough.
Whole ranks will be your “pocket.”

Not a bad thing.
This is the same as including tech for X deck that dominates.

It would actually make experimentation more interesting, and intricate.

As you should.
People have inflated ranks because of the game’s “generosity.”

But you still wouldn’t have an 80-90% win rate. The matchmaker is to ensure to lower that by putting X deck against you. (counter-matchup)

But it still wouldn’t be able to ensure it exact. If you are smart enough to work around it you’ll climb.

Your game plan is much harder to figure out by a system than the cards in it.

No, that IS a bad thing, as now you have 2 decks in a loop that only see each other and created a stupid 2 deck meta that probably becomes a 1 deck meta once people figure out which one wins that matchup.

That’s extra dumb because now if you like rainbow mage, the game is designed to give you the worst game experience possible with next to zero opponent deck variety.

And this applies to basically everyone else too that’s now trapped in their own tiny meta

People don’t have inflated MMRs due to the game’s generosity. Those are directly related to how well you are performing on a long term scale.

2 Likes

I don’t understand how you could think of a system that would accurately weight the powerlevel of a deck
You can’t get the powerlevel of a deck by judging just individual cards. It can give you a rough estimate on some base stats but you’ll be far from having the power of the deck
You can’t get the powerlevel of a deck by judging the synergy between the cards. You’ll have the obvious tribe synergies, but they can be terrible, it won’t give you a better deck than a top tier one with no tribe synergy. And for non obvious synergies it’s even worse

The only actual thing you can use to get a good estimate of the powerlevel is the place of the deck in the meta which requires that any deck can face any deck.
If you have a tier 1 deck that dominates all tier 2 decks, but that gets perfectly countered by a tier 3 deck, what is your system going to do ?
Because as I understand what you are proposing, those tier 3 and tier 1 deck will never face eachother.

Matching players based on their deck’s “powerlevel” means that you only allow brute force to be a thing.
If you make your tier 1 deck weaker to counter another tier 1 deck, then you will be forced against tier 2 decks that your modifications make you weaker against. OR you manage to face the other tier 1 deck that you now counter, and it’s them that are pushed in tier 2 and that you will no longer see at your level.

Not even mentionning the stupid question : how does a meta start if you need to estimate the powerlevel of all decks, but you don’t know what they are since you don’t allow them to play eachother ?

Now imagine all of this, but you include the fact that the powerlevel of a deck is dependant of the rank it is played in.
Pocket metas are already a thing in the game, but matching players by powerlevel of deck means that they will never see other decks being played at their rank. In order to see variety you will be forced to craft over decks. It could be a good thing that variety encourages variety but it’s not what happens here, since changing decks will make you play against players that already play against what you just switched to

2 Likes

You are pretending as if there is so much diversity in classes that this would a hard thing.

All the competitive decks could just be specifically placed to face one another, and if your deck in your chosen class doesn’t include the key cards that are expected to make it a competitive variant of a deck for the chosen class, then you queue into others not playing the competitive cards/decks.

More intricacy than this of course, but it truly doesn’t sound as impossible as you are making it seem. Other games do consider deck composition, so it’s not impossible.

You are taking my whole argument to the extreme.
Being more likely to face a deck to lower/heighten your win rate doesn’t mean it will be over 50% versus play rate.

It will just be more likely if you are overly losing against it. Same thing for a deck you are beating consistently, it will be less likely to face it to get you closer to 50%.

It doesn’t, I explained why.

You’ve turned this into being completely glaring. Other games that do consider deck power level to match you against decks on purpose, aren’t being that conspicuous about it.

Of course they don’t, but it also isn’t what I said to prompt your response.


Match you against the deck of similar power level which in this specific matchup is that tier 3 deck. Again, in a more likely, still “inconspicuous” level.

The matchmaker won’t be that narrow focused like “tier 1 good, tier 3 bad.” The variables needn’t be written here since I am not building it, but it’s obviously implied that it goes into specifics, and not just the face value.

You playtest your game like every serious company should.

No you won’t.
Again, obviously should I say, it will be more likely to face specific decks by considering deck composition but it doesn’t mean there won’t be variance… obviously.

Obviously, my argument isn’t that if you play a Tier 1 deck you’ll face a Tier 1 deck, it’s that you’ll be more likely. It’s just taking into account deck composition to the matchmaker, not completely dictating with it to the extreme.

We’re already on a level playing field.

But you still haven’t defined what better means, or how you measure whether or not your changes are working.

MMR is the same thing as ranking.

Making this comment demonstrates that you have no idea how matchmaking actually works.

Look at you helping me out, so I needn’t write it myself.

For this, you need machine learning/AI which has access to all the games being played in real time. Suffice to say, not even top IT firms have such AI access as of right now

Anyway, Derkan solved it. It’s impossible to do this reliably.

First of all, there’s balance issues.

If you’re aiming to have a balanced game with every deck at or near 50% winrate, then no true counter matchups even exist, and you’re always matched to a deck with the same strenght as yours.

If you don’t have such a balanced game at your disposal, then people queueing up the broken decks will always win more than their fair share, so you still can’t possibly balance the game by rigging it.

If you want to rig the game to be fair, you have to not rig the game at all. It’s just statistics.

what??!!?!?!

That sounds MISERABLE. Why would you do that instead of matching Smeet against another good player?