The reality is that the default of any state of any startup is failure. You probably won't understand the magnitude of the problems you'll face until you see them directly. At this point, your ideas are just ideas and you should work on validating them.
There's a lot of ways to go about doing this and it'll really depend on the market you're chasing. It's likely that a combination of customer interviews and sales will let you gauge which of your ideas have promise and which won't. Something that really helped me was thinking about questions VCs ask when they're evaluating a startup. What's your unfair advantage? Why hasn't this worked before and why will it work now? These questions are incredibly important even if you're not interested in raising venture. There's a lot of books and literature on the web so I won't get into this deeply.
You probably noticed that I didn't say writing code. Sometimes writing code is how you should be validating your idea but most of the time it isn't. You should get in that mindset. However, when you need to write code there's a few ways to get over that commute drain:
1.) Wake up earlier and put some time in before you start your day.
2.) Live closer to your job. It'll alleviate a lot of the problems of coming home and not wanting to do more more work.
3.) Write code on the weekends instead.
Also, whatever you do, don't do it alone. I did it alone for the first time and it's awful. Your job is a great place to find future cofounders or employees. Do hackathons with them on the weekends. Tell them your ideas. See what they excited about.
A year later, I'm doing some consulting on the side to pay the bills and getting into a pivot that's finally moving the needle. Even though most of this year was a total failure, I couldn't be happier and I'm glad I did what I did. Feel free to shoot me an email at jr.sunny at gmail dot com if you have any questions or need any help.