Game Masters do a lot of work. They are the ones doing the prep, adjudicating fairness, and roleplaying a whole world, while players show up to game day with a character sheet and some dice, there to relax and have fun. So, with Hostargo, I want to take as much pressure off of the GM as possible. But recent playtests have shown that is it possible to go too far in that direction.
One of the main mechanics I was excited for in Hostargo was the crash. When players failed an action, this was an extra die roll that they made to determine -by how much- they failed. This was a fantastically arbitrary number for myself running the game, as it gave me some input on how to describe the outcome of failures. However, its arbitrary nature also aggravated the players, because regardless of their initial roll, their character's skills, or the current situation, the crash roll was an independent determinate of their fate.
This simple but prominent mechanic has stuck around in Hostargo for longer than it should have. It had a clear negative effect on the game, but I was dragging my feet in changing it because it's purpose was really to make the GM's life easier. I couldn't really see, from only looking at my side of the table, how much of a problem it really was. It was under the players' skin. Not everyone could even pinpoint what they didn't like about it; after all, it made otherwise "nothing" rolls into "something interesting probably happens". After enough feedback, however, it has become clear that this (and some other mechanics like it), need rework.
There are plenty of ways I can deal with this mechanic. I can address its symptoms, or rip it up from its roots. The more important lesson here though, is that I need to pay closer attention to how each of these mechanics is affecting both sides of the equation. Especially since Hostargo is an asymmetrical game, there's plenty of room to ensure each mechanic not only does its job, but feels good to the people at the table.