As I spend more time coaching, I am more and more convinced that students best learn through games, not drills. I am lucky to have had an early coaching education that really emphasized games, but I can also see the weaknesses of that curriculum, most notably where it moved away from being game-based and became more drill-based. So now I'm a little bit obsessed with what defines games and how to turn those less interesting drills into games.

There's a book that came out a week or two ago called "Reality is Broken," by Jane McGonigal, which is definitely influencing how I'm thinking about this stuff. Her premise is that the reason that people become so obsessed with video games (and with games in general) is that they are doing a much better job of engaging people's interest than the "real world." Her question is what is so engaging about the games and how that energy can be brought back to bear on reality, which is sorely in need of that passion.
She breaks this down in to 14 premises, of which about half are useful in terms of the things I've been working on with the two clubs where I coach:

1. Tackle unnecessary obstacles - a major premise of what makes a game a game is that it involves overcoming unnecessary obstacles. For fencing, it's perfectly easy to hit somebody. It's harder to do it with obstacles - tiny weapon, smaller target area, time limits, space limits, priority or lack thereof, etc.
2. Activate extreme positive emotions - this is a boon for coaches. Playing games in class should make coming to class exciting because games themselves promote positive feelings.
3. Do more satisfying work - give clear goals and vivid results. Fencing should be great for this. The result of hitting somebody is pretty darned vivid. But when it's not a game, when it's just a drill that ends with the same fencer always making the hit, then making that hit doesn't feel like a win, which is a waste of that jolt of positive emotion.
4. Find better hope of success - games make even failure fun (in video games they give entertaining animations when you die, for instance) and people learn best by making mistakes and then fixing them. Also, games give attainable goals in several ways. Advancing from one level to the next is a clear path, with lots of early achievable milestones, then fewer milestones that feel like more substantial achievements. Martial arts schools do this really well. My old club did an okay job of it at first, and one of the major places where we lost people was when they ran out of clear goals. Putting these ideas together gives the idea that even if you fail this time and the next ten times, you will still succeed at some point, which is satisfying.
5. Strengthen social connectivity - this should be a no-brainer, right? We're creating a community of people who have things in common, like a shared interest in games, in combat, in athleticism, etc. One of the clubs where I coach now, though, a university club, has very little social fabric. I've never encountered this before. How on earth can a university club not be all about partying together? And yet, there's nothing there, because the club thinks that its only real goal is to teach fencing. But if people don't show up (because they don't meet social goals through the club) then they don't learn to fence. So many clubs think that their only role is teaching fencing, but that's really only the least of our services.
6. Immerse yourself in epic scale - Harder for fencing, but perhaps one of the reasons why we are so loathe to give up allowing everybody access to national events. We like the epic scale of our sport. This is actually also influencing one of my non-social attempts to fix the university club - a large-scale achievements game to encourage involvement and competition with all team members (as well as show individual successes and areas for further growth).

There are lots of other good ideas in the rest of the book, but the above ideas constitute the first third or so and are the most easily applicable. I'm definitely taking them into consideration as I plan my games for each night's classes and as I build the larger ongoing game for the university club.

I would never have though of myself as the type to end up reading lots of books about game design - I end up playing a round or two of Solitaire every few years whether I need it or not - but I think that that's my new direction for extracurricular reading. I'm very much interested in what characteristics make games fun and addictive and how I can harness those feelings for coaching.