To that end, there's a certain value in declaring that, "You know what, PyGame Zero's event loop is probably fine. Let it sweat the details, and we'll worry about writing the character.jump() function next." Another framework I love that works this way is Love2D, which targets Lua but has a similar approach:
Sure it's not suitable for any project of decent complexity, but there's a reason we keep reaching for it during Ludum Dares, and any time we want to rapidly prototype an idea. It takes the least amount of effort of any tool I know to go from concept to prototype, and get the idea in playtesters hands so we know if it's worth doing properly. Getting to do that in Python sounds awesome, so kudos all around; this is neat stuff.