No not silly at all 
Ectocomp: Any time spent planning, thinking, doodling, researching, etc. DOES NOT count.
You’re allowed to plan for ectocomp, and if you’re going for the speedwrite I highly recommend you do. I usually know exactly what the plan for the game is is before I start so I can just start writing rather than worrying about getting stuck for ideas part way through. You need to try to be disciplined and not let scope creep happen. The story you plan is the story you write.
It is a bit of practice. I think one of the reasons why I like the petite mort so much is I’ve enjoyed writing some short story and flash fiction here and there over the years and it feeds into that. But you can start anywhere.
Best advice I can give is keep it small and contained. Think about one event or one day for example you want to write about and go from there. Keep the cast small and manageable. If going with a normal choicescript game format, go with a tight introduction, story (branch and return, or gauntlet with early ends), multiple endings. Consider leaning into some “mood” setting “fake_choices” that don’t necessarily require entire branches for a choice. The “next” button is also a really good spot to replace the “next” text to put up to a few words “impact” text in a spooky story. If you take a look at previous years you can see the types of games that come out though, there’s usually some parser and twine games, mixed in with the odd bitsy, ink, choicescript and original coding which all have their strengths and weaknesses for making short games with.
Alternatively you can go modular. I did that one year where I had the “essential” scenes I needed to make the game work from start to finish, and some “add on nice to have” side quests/extra scenes I could add if I had time. (I did end up having to leave some out.) The downside I had with that was my stats weren’t balanced quite right, because I left a parts of the game I’d potentially planned to put in out and I ran out of time to figure out how to fix it, so I have mostly avoided doing that in other years. (For less stat dependent games it’d be an option though.)
I recommend overestimating how much time the writing will take so you stay under. (Ultimately you do need to guess. Estimate based on other writing how many scenes you could write in 4 hours if you have the story pre-planned in your head. It’s going to be different for everyone )
You can break it up to make it easier. It doesn’t have to be a 4 hour marathon. Just record the time every session you sit down to work on it.
And finally, don’t stress too much. If you go over time, the worst that happens is you can enter it in the Grand Guignol section (more than 4 hours development) instead. I’m actually going to do that this year, because I went a bit overboard editing, adding sound effects, customising buttons etc on the game I made. So just have fun with it and see what comes out 
)