To be honest, I couldn't really make heads or tails of this game at first. I reread it a few times, went and read the original rules for SET and then reread the game again. I understood the mechanics by that point, I just couldn't figure out why the players would care to match the sets when they would just be observing the clues in world. The thing that wasn't clear to me is that each table is meant to be observable traits, and traits that are hidden until the combination of observable traits are discovered. From there, you're able to start to logic puzzle out the other potential options based on the SET mechanics. That's a really interesting way of handling investigations. I do wish that the intent behind the specifics of the mechanics were explained a little more clearly. 

I think it also might have been helpful to just leave out the rolling for the sets at the start and just instruct the GM to choose. It took me a while to realize that the instructions to roll 2d20 and only reroll the maneuver-flaw is to just increase the likelihood of getting a valid SET as the player's target. I don't know that the randomness really adds much, since the SET system already abstracts the possible combinations to some degree.

Over all, I think this game is pretty interesting. The system of generating and structuring clues is innovative. Its a well that I could see myself exploring in the future. Thank you for that. I'm always happy to be exposed to new ideas.