[MENTION=6801204]Satyrn[/MENTION] - Gonna allow 2 maneuvers on a nat 20 for sure. I agree that it's gonna need some play testing to be sure, and my player will agree to adjust if it proves necessary, but would still like to get at least close with the initial array. Gonna attempt some math now.
For CR=PC Level opponents, I'd like to keep the odds of a maneuver around once per four rounds. This is a little worse than the 4 per 10-12 rounds that the short rest mechanic grants, but should be made up for with lots more maneuvers against lower CR enemies.
12,9,6 DC array, PC will have a +1, +0 and -1 in Wis, Int and Cha respectively
At 3rd level Against a CR 3 monster, he would need between 14-16 (25%, 3 per short rest)
At 7th level,Against a CR 7 he would need between 15-17 (20%, 2.5 per short rest)
At 15th level, against a CR 15 he would need between 20-22 (5% .5 per short rest)
That's good at 3rd level, but I think I need to relax 7th and 15th.
12,8,0 DC array, PC will have a +1, +0 and -1 in Wis, Int and Cha respectively
At 3rd level Against a CR 3 monster, he would need between 14-16 (25%, 3 per short rest)
At 7th level,Against a CR 7 he would need between 14-16 (25%, 3 per short rest)
At 15th level, against a CR 15 he would need between 14-16 (25%, 3 per short rest)
Now, it is going to get more difficult hitting level equivalent CRs, fore example at 6th level against CR6, he'll drop to a 10% chance, so the question becomes, should the DC instead be a sliding scale based on fighter level. Something like 15+CR-LV? That would make the power curve steadier, making it so that the base odds of the maneuver are always 25% for a foe of CR equal to level. That said, given how often he'll be able to use maneuvers against lesser foes, I might be fine with him having trouble with more powerful foes, "Your feeble skills are no match for the power of..." And the nonlinear progression also more strongly encourages him to invest in some ABIs that he might otherwise use to take a feat, which I kinda like.
On the flip side, not being able to decide to use a maneuver at a time of your choosing is a significant loss, so it might actually be okay to go in the opposite direction, allowing for more maneuvers since when they happen won't be as assured.
Gonna think on this a bit more.