When I first designed Shadowitz, I made it with a mechanical push toward specialization in methodology as well as skill training. I wanted something that promoted a particular style. The aggressive runner did well with straightforward combat tactics, but also with straightforward social interaction, while the sneak made a better sniper than infantryman. As we've played more, that's becoming evident to the players, and one of them in particular (the one who actually read the rulebook when I finished it), has really embraced this idea of describing what he's doing in ways that let him leverage his strengths. I encourage that kind of flexibility, because it gets the players really thinking about their characters and situations; it's harder to say "I just roll" when you're figuring out how to make said roll the best possible, and doing so requires engaging the game narratively.
With the players more carefully considering the how of their actions, and me still making suggestions along the way to show how I envisioned it working, the players very rarely find themselves rolling pools with a method lower than 3. The hacker/rigger/face/adept still gets stuck with 2s sometimes, but he's spread very thinly, so sometimes he's just got no choice. Most of the time, however, players are rolling for 3s or lower, and shooting for less than 5 a large part of the time. Combined with the fact that they like to roll the knacks that they're good at more often than not (and who doesn't), it means they're coming up with a high number of hits every roll.
That's good. I want to reward them for playing to their strengths. But it also means that the current hit thresholds I have for the game are kind of ridiculous. For an average person, they're fair, but they make my team near superhuman. For example, I have hard at 5 hits. Between gear, cyberware, adept powers, and aspects, any member of my crew can score 8 hits in his area of expertise without even trying. Easy is 1 hit, hard is 5, and they're clean off the charts nearly every time.
My kneejerk reaction is to simply double all thresholds, making easy 2 all the way up to making hard 10, but that makes hard rolls nearly impossible to pull off. Even someone trained all the way to 10 in a knack with a 5 method won't make it most the time (though with aspects and gear he probably will, since those can push the pool over 10). So, rather than do that and leave it that way, I think what's really called for is the creation of a more robust difficulty table, that defines difficulties for a larger range of results. This lets me keep easy in the range of the possible for those times when the runners need to roll something outside their area of expertise, while making the difficult things difficult for them. It'll take some tinkering to get the balance right, but the first order of business is to jack that high end way up. No more 10 impossible things before breakfast anymore.