This leads to 3x 2x and 1x of certain cards every time. It is extremely unlikely they will change from this model. Yes the 2x cards are stronger and the1x cards are usually the strongest. However 3 core sets for a full playset is usually the same cost as a single booster box of other randomized card games. You still save a ton of money and get a playset of the core game.
Except it doesn't lead to that every time. As I pointed out, Call of Cthulhu doesn't use this model, the Core set has exactly 1x every card - no duplicates at all. I am hoping they will return to this model which they have already used in the past. It may not be a deal breaker for everyone, but it will be for some and for others it will dampen their enthusiasm for the game, potentially causing them to evangelize it less. I know I would be less happy if I had to throw away a big pile of duplicate cards. You're saying you wouldn't be? That if I told you that you could end up more unique cards for the same price that you wouldn't be interested in that?
Also, it doesn't matter what a booster box of a totally different type of game costs, that's irrelevant. I don't play any CCGs specifically because they are money grabs, that's what's so great about the LCG model.
Comparing to something more relevant, look at monthly pack costs. The first Core set is a good deal. $40 is getting you 252 cards compared to $15 getting you 60 cards in the monthly pack. The break-even point is about 160 cards. The second Core set may not even reach that number if it's the lumpy distribution model. The 3rd pack is definitely a bad deal, you're only getting any value for the 1x cards, and there likely aren't enough of those to make it a reasonable value per dollar ratio. $40 for the final 20 cards as in Surreal's example would be quite annoying.