You’ve pinpointed one of the issues with balancing the meta, which I’ve already mentioned above: from the very beginning, instead of fixing ‘broken’ cards, archetypes, mechanics, interactions and so on, they’d ‘release’ these so-called ‘hate cards’, supposed to counter them specifically. I’ve always been against this design approach, to be honest.
Nothing personal, but that’s what I referred to as being a ‘metagame hussy’ above, or, more precisely, bad game design in which being one is the winning strategy.
An extreme example of that would be a game of rock, paper and scissors. You know somehow (with some information from outside of the game, which is literally the definition of the ‘metagame’, or shortened to ‘meta’, as opposed to what some people assume) that ‘everyone’ is playing rock, so you go with paper instead. In time, the popularity of paper rises, so some ‘clever’ players switch to scissors. One more iteration, and ‘everyone’ is playing rock once again. There are people who fancy themselves as experts, ‘meta analysts’, as they’d put it, writing ‘meta reviews’, as if it were some kind of science — and they’ve even invented a term for what’s described here: ‘meta cycles’.
The sad truth hidden behind these big words, however, is that the gameplay itself is more or less meaningless, as are a player’s skill or decisions in the game, and the most ‘successful’ one is the most flexible — or lucky — ‘meta-hore’ who happens to ride this tide just ahead of yet another wave crest, so to speak. This is indicative of a bad design — just bad and unfun, even if it’s perfectly balanced, as ‘rock-paper-scissors’ are, mind you, which isn’t even always the case.
The opposite pole would be chess. Sure, you can use some metagame information, e.g. what kinds of positions or openings your opponent prefers or plays best, to prepare for a match, but generally, the outcome of any game is mostly determined but what each of the two players does at the board and their decisions during the game, not by what the bulk of other players prefer to do. I think this could be achieved by good design, the aforementioned ‘hate cards’ being a typical counterexample of it, with a decent level of strategic depth and diversity of playable archetypes, decks, cards and so on. The closer you are to this, the more you’ll realise that every competitive deck that you’d face could be an ‘off-meta’ deck, at least ideally.
Sure, it’s easier to just cobble up a few obvious deck archetypes for a rock-paper-scissors-like scheme (insert your ‘aggro’, ‘control’, ‘combo’ or ‘midrange’ instead, and here you are) for a semblance of a ‘diverse and balanced meta’, but it’s less fun, although it’s probably cheaper to make and more profitable to sell.
PS One more thing: instead of refining a game, like chess has been, resulting in a masterpice of game design, much appreciated over centuries, it’s apparently more advantageous for them to appease the false idol of 'nEw CoNtEnT’ 
and keep selling their junk at regular intervals. See also this on the subject (yeah, repeating myself again), including why this fetish is not always even good for a game, Chess 960 being a prime example.
Despite being such a masterpiece, chess is so hard to monetise, yes, there’s that.
Thanks.