I wasn't really feeling it at first to be honest. The mechanics as they are seem to encourage a gameplay loop of standing still for long periods of time until something relevant happens, then passing a basic but frustrating timing/reaction check. I think the main problem is that a bash requires too much investment. Not just the charge, but also the execution time. Since the jumping part grants neither speed nor iframes, the delay between when you release your charge and when something useful happens is usually longer than it takes a freshly launched tile to reach you. This combined with the attack's brief hurtbox means you have a very small launch window just before the tile starts moving, which is very difficult to hit since its telegraph is so vague (when it doesn't just launch from off-screen). It also doesn't help that there isn't a clear indicator that a full bash is ready, especially since a non-charged one is pretty much useless.
However, I then found out I can charge my next bash while dashing and much more interesting strategies became possible. For example, using an initial bash to clear an opening in a large group of tiles, then kiting the others with a roll while charging another attack. This discovery didn't alleviate the above problems for me, but when new strategies emerge with player experience, that's a sign of a good core design, which in my opinion is the most valuable trait in a game jam, not mechanical issues that would probably be solved with more time.