The meta is okay. There are three classes with good performance (Rogue, Priest, Druid) each with a totally different archtype (tempo, control, and midrange). And occasionally you'll see a shaman, warlock or hunter.
I don't really like playing with or against the Rogue deck, mostly due to the highroll variance. Prince Keleseth + Shadowstep on turn 2? You win! Oh you drew patches and didn't get Keleseth on two? You lose! Apparently you get Keleseth and a Shadowstep around 10% of the time if you mull for them. And when you get them neither your skill or your opponents skill matters at all. Because that opener is just so strong.
Which means of the top three classes every single one has a win condition which feels horrible to lose to because you cannot really do anything about it. Too huge never-ending Jades, losing to priest machine gun fire, or losing overstated charge minions.
Almost as bad as when you used to lose because you were at 28 life on turn 10 and miracle rogue could do that much in a turn with no board.
Yea, got into a discussion on a youtube video about this. The meta is 'balanced' from a numbers perspective, but not from an interaction standpoint.
So many decks are just disguised solitaire. Who gets their key cards, or not, determines the outcome. Rather then your in game decisions.
Flipping a coin is 'balanced', but not fun. You will win roughly 50% of the time, and you cant affect that. Whenever hearthstone moves too far in that direction, it isnt fun. The devs dont seem to understand how to design cards that are interactive on a broad scale without being OP.
They make narrow hate cards that almost auto win against the right deck, and almoat auto lose against other decks. Or, they make cards that are so flexible and powerful that they go in everything.
The base game rules make it harder to create interactions, since your opponent has almost no disruption during their turn, but its not impossible.
There should be ways to interact with any card type. The design of deathknights was flawed in that aspect. Even if it was just a card that punished repeated hero power use. The 'hate' cards, in general, need to be better against the field and weaker hate effects. I.e. eater of secrets should be like 4 mana, 4/4, destroy all enemy secrets. Period. No stat buff per secret.
Its not as useless against non-secret decks, and not as auto win against secrets.
Hungry crab- 2 mana 2/3, destroy a pirate to gain 2 life.
Blood knight- 3 mana 3/4, destroy a SINGLE divine shield to gain +1/+1.
These examples arent directly related to the top decks at the moment, but illustrates some of the issues with their design process in general. All the hate cards dont really get played, unless the deck they target becomes the #1 deck by a fair margin. Even then, lets say the big deck is 50% of the field. It feels unfun to draw the hate card against the other 50% because its often worthless since its so narrow.
If we had a card like: 5 mana, 4/5 dragon, when opponent uses a hero power, deal 3 damage to a random enemy character. Thatd help against razakus, as they either have to hit it 3 times with hero power (taking 9 on an empty board) or have some minions out to trade with it. Dragonfire doesnt clear it, and neither shadow word removes it. Using it at the right time can throw a wrench in their plans, or at least a speed bump. Similar to what loatheb used to do. But, its decent against other classes too. Mage cant ping it profitably. Hunter cant just finish you off with hero power. Paladin cant really use hero power on an empty board. Its not great against rogue, but thats ok. It shouldnt be great against everything, but a 4/5 for 5 isnt the worst. Overpaying 1 mana to make it awkward to re-dagger for a turn is 'eh, fine'. Its not winning, but its not so bad that you lose cause of it.
Anyway, the point is we need more decision points and interaction overall. However they attain that is fine with me.