Make Lots of Games
If you make one game, and I make ten games, you could say I have a ten times greater chance of being published.
Sure, your one game might be better than the average of my games, but I still probably have a five times greater chance of being published. This doesn't mean I complete all these games. I abandon most of my games early on.
One game to rule them all
Some people seem to think that the way a great game is made is by working on one idea endlessly, until it's great. They imagine that anyone who's sufficiently industrious can therefore make a successful game.
If you only ever make one game, you'll never make a great game.
There's so much that needs to be learned, and it can only come from a variety of experiences, successes, problems, and failures. Spending all your time on one game doesn't give you that necessary experience. Even if you did become an excellent designer later on in the development of your game, it would be too late to apply this knowledge to the important early choices you made for your game.
If you made a second game, wouldn't it be better than your first game?
Where I went wrong
When I look back at my early failed prototypes, I can see mistakes in them, that I couldn't see at the time. As an example, I didn't understand goals properly, until I made my 37th game. That alone probably killed most of my early games. There are likely all kinds of things I still don't know.
Game design is really hard. I don't think anyone really knows exactly how to make a great game.
My publishable games have not come from some stroke of genius, but from a lucky choice of theme or mechanic, plus a lot of skilled work.
I expect this is the same for all creative works.
Just keep trying different ideas, until you discover a good one.
Things go wrong
I made a pirate game. I put a lot of time into it. I couldn't get it published. Was it a bad game? No. Was it about pirates? Yes. I didn't realise how hard it was going to be to get a game with no new theme or mechanics published. So, it went back to my shelf.
A friend spent years making a game clearly evoking the Power Rangers, in everything but name. While we were working on the game, the company that owned the IP suddenly announced a series of Power Rangers board games.
I've seen many people's "masterpieces", and just thought to myself "clearly this isn't going to work", or "There's really no audience for this." That person might be you.
When I play another designer's game, I rarely quibble with the details. The game has huge issues, that the designer is completely blind to. Even if it's good, the game needs major structural work. Other games are feeble, and unsalvageable.
It's just not great
Even if you don't make any mistakes, it doesn't mean your game is good. The central idea of the game might just not be that good. And, even then, there's no value in a good game. You're looking for a great game.
There are so many mistakes and misfortunes that can befall you, that a single game has a good chance of being doomed, due to outside circumstances, or due to flaws you cannot see or understand.
Make lots of games.