Sometimes I can make a game in a month, then sell it to make $1000 or so. Maybe that's not great by most people's standards, but if I could do that every month I'd be pretty happy, that's enough to live on.
But most of the time it doesn't work out so well. Making games quickly requires quite an intense level of effort that isn't really sustainable long-term. Sometimes I get sick, and while that might not prevent me working, it limits my potential. Sometimes things break; computer hardware isn't very stable under immersion in liquids or high-speed collisions, and repairs or replacements can be pretty costly. And sometimes I'm an utter fool and write over a month's work that I haven't backed up.
Running some disc recovery software now and hoping it can rescue things.
And other times, I can spend months prototyping various ideas that sound cool in my head but don't actually work. Some simply don't work at all, but others seem quite close to working, if I can just think of the right way to bring their pieces together. Trying out new ideas takes a lot of time and gives a lot of failures. I can see why some prefer to stick to variations on an established formula; trying out new things takes a lot longer and tends to be less favourably received.
When I make a game in a month and it works I feel pretty amazing. I feel like I could do that every month no trouble. But I can't, because there are all these hidden costs that I don't perceive at that point in time. Those months are a few among many. Just counting the development time of a game from when I started working on that particular game until when it was released misses a lot: the risks of accidents happening that didn't happen to occur in that time, the time spent on failed games that mapped out the way to that one, the time spent fixing bugs and trying to convince anyone to review the game afterwards, the subsequent period of exhaustion from overwork. It's not enough for a game's sales to just cover the time spent making it.
I'm very lucky to be able to continue making games for now. Lucky to have had a game on Steam and the Indie Royale bundle, that actually made enough to pay off my student loan and live on for a bit. Extremely lucky that my wife has a job at the moment that pays enough to support us both. I feel kind of bad that I sometimes complain about not having much income from games when there are others in genuine need. But if there comes a time where I need to choose between trying to support myself entirely from making games or going out and getting a real job this is something that I'll need to remember: I can't consistently make games as quickly as I think I can, I have to take into account these hidden costs.