I'm surprised nobody in this thread has actually run numbers. I don't care to because I don't think its a problem, however, here's how if someone is so inclined:
You have playtesters listed on the backs of every expansion pack. Get all the tester names from the waves about a year ago and see if there's a filter in ListJuggler for the last year of tournaments. I think I could see people up in arms if every single one of the listed names showed up in the top 25% of every tournament they entered when the thing they tested entered the meta. I suspect that is exactly not the case and that good players are good players.
And of those non-playtester good players that do very well I would also suspect that the second anything is previewed they are either at home or Vassal testing it out and adapting their current strategies to new lists and upgrades. I can see how certain cards change things subtly enough that mathematics and probability can play a role in determining value, but we are still talking about lists based around pillars of list construction. There's not a whole lot of those. Experience will adapt your playstlyle and list to those subtle changes in card economics.
Edited by jonnyd