Poker is a very different game, where you play many rounds to determine the outcome of a single game. That evens out the RNG within one game. On top of that you can fold a round, increase the stakes and bluff. There is simply no comparison.
Firstly, I did not say it was one to one, I said the opposite -- "let me take it to the extreme" (implying the closer games like MTG also successfully play with randomness). There is a lot of comparison to poker though: Random draws, reading your opponent, playing optimally.
You don't need to fold/raise etc however, in this game, the best players rise to the top. How can certain players in Gwent consistently play near the top of the leaderboard? It's not random, even those with hundreds of games do not just appear at the top of prorank. I watched BeardyBog's stream last season when he climbed and held #1 -- there was a lot of thought going into it. How much skill was involved for LordBushwook to win the tournament (forgot the name now) twice in a row?
You're not going to win each individual game, but in the long run, you will win most, this is fine in a ladder system where you are playing hundreds of games. By the law of large numbers, you'll sort yourself out to your correct MMR. This is valid.
What about in a tournament setting? In the round robin setting you play 6 or more games. In the elimination bracket, you play a series best of 5 at minimum.
Let's look at best of 5.
If a player has a 5% advantage (in terms of likelihood of winning over his opponent), he will win the series 59% of the time
10% advantage -> 68%
15% advantage --> 76%
What about if a complete new player faces a veteran (say the veteran has a 95% chance to win, in the very small chance that his opponent outdraws him and win 5% of times). In a best of 5 series, the veteran will win 99.9% of the time.
Note that if we change to best of 7 series, each advantage is much more impactful (.61, .71, .80, and 1.00 respectively)
This holds true RNG or not, and I get that the post is about each individual game itself, but my point here illustrates that the better player DOES win more often in the long run (which is also true for any game).
However I've already explained before how each individual match does not rely heavily on randomness.
As a personal anecdote: I've never lost a single game vs a new player in Gwent. I in fact always feel bad that they are matched up with me, since I can see every mistake they make, and can often finish the game with massive card and/or point advantages. This is not the mark of a game where randomness rules.
You're free to have your own opinion that beta gwent was superior of course, but I just haven't seen any compelling evidence or argument to convince me that the current version of Gwent isn't a much healthier game with a better design space for future iterations.