For the last 9 months or so, I've been taking classes at a fencing school here in Vancouver. Back in undergrad at the University of Colorado, I did a few years of kendo which I rather enjoyed and wanted to seek out something similar up here. What the school teaches isn't Olympic sport fencing though, it's Renaissance era swordplay. Basically drawing on all the sword masters mentioned in the fight between Wesley and Inigo in The Princess Bride, especially Capo Ferro (although the actual fight choreography in the movie has almost nothing to do with what those masters would have taught).
Having practiced for a little while now, it's interesting noticing some game-like properties of what we've been learning. Being that it's not sport fencing, there aren't strictly defined rules per se. The "rules" (beyond the stipulations explictly for safety's sake) are just the techniques that work well and help you not get stabbed as much. But there are still things that emerge out of the constraints of what your body can do and the desire to not get stabbed. Obviously a duel-like swordfight is not dissimilar from fighting games, given that's what those games meant to simulate anyway. Some of things I've learned in swordfighting are directly analogous to things in 2D fighters, but they can apply more broadly as well.
In Capo Ferro's fencing manual, he discusses a notion of "tempo" which basically describes opportunities to strike one's opponent. There a four tempos described by Capo Ferro.
First Tempo (primo tempo): As the opponent is first entering range or generally before they've taken any action.
Half Tempo (mezzo tempo): During the preparation of an opponent's action, our attack finishes before his can become threatening.
Counter Tempo (contratempo): As the opponent attacks, we respond at basically the same time (but with a slightly faster action and superior positioning or strength).
Two Tempos (dui tempi): An attack that takes two actions, the first of which constraints or inhibits the opponent somehow.
In the context of video games, obviously things like attacking an enemy while they're in the wind up for an big attack is something that's been done for ages. But these same actions can easily exists on a high level of tactics. A contratempo response to an advancing enemy force in Starcraft II would be to force the engagement onto a ramp where your forces have a superior position. Or to allow your forces to engaging their main attack host, but quickly slip Stalkers into their resource line, crippling their economy and rendering any possible success of their main attack moot.
One of the common challenges I find in design is finding ways to provide good structure to the the game's mechanics. Structure provides patterns and models, and those make mechanics easier to teach and learn. An action or strategy game that provides meaningful ways to act in all four of those tempos would probably have enough depth to be interesting.
I suppose the broader point is find ways to structure your game's mechanics and be open to seeing structures in places you might not expect. That structure could prove to be tremendously valuable, not just for developing your game's mechanics, but for conveying them to someone who doesn't possess intimate knowledge of them (and this is honestly one of the biggest design challenges for any game).
And if you live in Vancouver, come check out Academie Duello. You can try any class you want at no cost. Certainly more enjoyable exercise that hefting metal disks, at the very least.