I've been thinking on my player's argument last session that trying shouldn't make things worse. Or at least, that trying in very bad circumstances shouldn't be more likely to make thing worse than to succeed.
Giving sufficient penalties - say, a -4 or more for being grappled, plus -10 for darkness, plus -4 for posture, can mean you can't really do anything at all without almost certainly critically failing. With a -18, a 12 skill plus an All-Out Attack (Determined) and Telegraphic Attack gives you a net 2, so you succeed on a 3-4 and critically fail on a 12+. You need an 18 skill to get to an 8 or less with those two.
I think my players would say they're not opposed to penalties, just to nonsensical situations where trying makes something worse. But trying often makes things worse in real life. In my experience in combat sports, I've learned that doing the wrong thing hard is worse than not doing anything, and often, your opponent is putting you in circumstances where they need you to try to do something to allow them to continue. Sometimes doing nothing is the best move - the cost to you to do nothing exceeds the cost to the enemy to keep doing what they're doing. A classic example is a grappler who is expending energy (FP in game) and position (spending CP) to try and finish a move. If you can defend it, you're better off just waiting them out. Or in our last game, a giant frog trying to bite your head off, unable to let go, while your friends come to help you.
I think the only way to make this incorrect, and make it so trying is always worth it, is one of two ways:
- a hard cap on penalties.
- removing the 10+ failure means critical failure approach.
I don't think the first makes any sense without a hard skill cap. If you cap penalties, you must cap skill (not bonuses, skill) so you can't just outspend your potential penalties.
I don't think the second passes a reality test - if it's possible to have a higher chance of a spectacular success than a spectacular failure, then the reverse should be true. And catastrophic failure shouldn't been too limited, otherwise a merely hard task is effectively the same as a truly impossible task. Plus, it means that there is little problem with repetitive tasks as you can't really make it worse.
So that's why I think that the rules as written approach makes good sense and reasonably models reality enough to keep as-is.