Spare Time Development and Polish?

I've ​been working with Blur very on and off again for months. All while planning future gameplay systems, an ideal aesthetic, and an environment for this player perspective to navigate. Though all of those ideas are still just scribbles in a  word document. My work in Unreal has been mostly focused on the core movement player experience. Where I'm naturally very excited by new design/mechanics to implement, I'm also hyper aware that the finer math and blueprint work towards a better game-feel is what I should still focus on.

After each few weeks hiatus of work, I'll need to pour myself back into the blueprint and remember how it worked and what I wanted to change. Though once I'm in those weeds, I properly feel the momentum and motivation to keep working until I know I've accomplished something. Whether that be implementing a new system or fixing part of one to work/feel much better. Though something on hindsight I can tell was getting me too deep in the weeds was the fine control of the limited look perspective. I still aim to work on and tune it, but decided to compromise with myself and implement ancillary systems still designed for immersion of the first person experience. These namely being foot step sfx, head sway, and sprinting. The sprinting I've had for a while, but paid significant more attention while coding the former systems to depend on it. This so the SFX and head sway would transitions between of "Walking" and "Sprinting" versions seamlessly. The current code is based off the player velocity to make said transitions and ensure the systems won't fire while walking up against rigid pieces of the environment.

In broad strokes, it all works as I want it to right now. Though, I again feel the pangs to tweak values towards feeling more natural. Like how the head sway currently doesn't match the foot step rhythm. I'll definitely tweak that, but I think another lesson hit home while finally finishing my footstep system. I was having trouble making sure it would reset properly as every possible transition happened in the code. While I did finish and had it working just as I wanted, I decided to share my gruesome spaghetti code to a programmer friend of mind (largely to disgust him). He was disgusted, but he also offered to help me rewrite it. It took what little was left of the night, but he completely condensed it to work elegantly. Leaving me to lament not being smarter than I am, but also wonder if I should really keep myself so bogged down in pre-polishing. Now largely aware that any system I write will likely be cobbled together and inelegant, should I really bend over backwards to make it do EXACTLY what I want before a programmer can even look at it? I KNOW how important systems polish is for quality and KNOW I'm better off ensuring it's done now over forgetting until it's too late, but...

this is a  p r o t o t y p e  to see if my design concept even plays well. It's forever satisfying to code a system closer and closer to how I see in my head, but it is still just walking. The intended player experience is more than just this walking. So for the sake of actually determining how this "90 degree look" plays with horror game systems, I think those systems will be my next major focus.                               
                                              ..AND/OR an actual environment to explore..
                                                                                                                                                      ...no promises