What follows is rambly. You’ve been warned.
I’ve been thinking about another result of the most recent playtest. It was implicit, between the lines of what John said, but I think the lesson was that the game needs a bit more space to breathe.
The game has two economies, cards and tokens, that interact. It’s generally the case that you spend a bit from each in every scene, or at least every scene where you get or take some spotlight time. You get them from, you guessed it, certain moments in the spotlight. This means that the diastolic and systolic elements of the system are very tightly coupled; every moment of gaining currency depends on spending currency, and every moment of spending currency can lead to gaining it.
This can suffocate the story in between. It makes it easy to keep your eye on the mechanical pieces, the currencies, and not think about the story effects you’re making and getting as you do the mechanical game. It makes it easy to evaluate actions in terms of net mechanical effect over embodying a character.
(Yeah, there’s the other extreme, where you’re left drifting between mechanical reference points.)
One way to counter this is to divide scenes or beats (or whatever your relevant unit is) into upbeats and downbeats, action and recovery, spend resources and gain them. This moves the systolic and diastolic rhythm into being a first-class element of your game. They don’t have to alternate by any means, but they should have some dividing lines. Consider Fiasco‘s white dice and black dice: they divide scenes into some concept of up and down (about outcomes, not actions, but still), and let you pace them as you feel necessary.
Another option is to minimize the presence of “resources” and “economies” in your design, as you see in Apocalypse World.