He doesn't mention how many hours he spent marketing it. I suspect not that many. I fully sympathize with his struggle, but this is a trap most programmers fall into so easily. Market need/fit > Marketing > Design > Programming when it comes to software products. You can't just build it and expect that they will come, unfortunately.
In "Start Small, Stay Small: A Developer's Guide to Launching a Startup," Rob Walling states three wrong reasons to launch something: 1. Having a product idea 2. To get rich 3. Because it sounds like fun This and other books on launching your own thing stress that you first have to find a market, then fill that market's needs. In the world of indie games I can't even imagine what that would be.
You can engineer a relatively high chance of success but it requires two things:
1) Actually picking an under served market. Meteriodvania style games are not one of these, there are a ton of game out there fulfilling that need.
2) Really really understanding the audience. You have to know what details of the previous games in the genre were actually important, and what were not which you can innovate in. Even the tiniest details can be super important.