Would people really pay for crap made in game maker?
Hell i was surprised when that RPGMaker looking game call of chthululululu sold on steam.
I wouldn't pay for "crap", but I would pay for an enjoyable and well-made game, even if it was made with Game Maker.
To be honest, I've never seen a Game Maker game that I'd pay for. But, to be fair, most Game Maker games are made by people that aren't trying to sell. Money motivates. Therefore, it doesn't exactly seem logical to judge Game Maker on fan-made games, or even any freeware Game Maker game. In some cases, I've seen people trying to sell GM games on their own websites, and even those sucked, but it was obviously not Game Maker's fault, it was the developer's fault. The problems were in the graphics, the story, the mechanics, etc. It's clear that they didn't know enough about using Game Maker properly and efficiently.
And it's not just Game Maker games that can suck. I've seen my fair share of other crappy games that were programmed in a "real" programming language.
Point is: It's not completely about what you use to make the game. You've got to have the right skills and the knowledge to make a good game as well, or your game will fail no matter what language it's been coded with.
I really wouldn't get your hopes up though guys, steam is notoriously restrictive over what they allow into the store, your only real way as an indie is to already have an large established community around the game - as most indie games on steam have.
Send a message to the steamworks support team, and they will give you a brief overview of the TRC they require before releasing games on steam.
No one should ever hope to get anywhere if they don't work hard. I plan to work hard to get my game on Steam. I'm making my game with Game Maker to speed up development by a few years, since I know GML fairly well and it would take me a very long time to learn a more professional coding language.
That said, I don't plan on making use of Game Maker as most people do (as I also have in the past, even with the Spirit's Quest engine). I wouldn't expect to get far at all if I did. A good example of poor GM use is how common it is for people to place tons of solid collision objects within a room, when usually a simple 2d array of "solids" can take care of all those collisions. Cutting down the number of objects is a great first step in using GM efficiently. I go by the rule that "Less objects is more". Another thing most GM users overlook is proper loading and unloading of resources. The trick is to only store what you need in the game itself. If you must, just encrypt the external files (i.e. sprites, music, sounds) and load them and unload them at the proper times. This decreases the initial loading time immensely, and it should free up a bit of memory as well. The biggest flaw I've found in GM games, including mine, is that they are needlessly bloated. If GM is used right, I think it's very possible to make a professional feeling game that runs fast, looks good, feels good, is fun to play, and that will ultimately sell well.
As far as having a "large established community" around the game: I think that, if your game is good enough, that will happen mostly on it's own. Of course you need to get your game out there in the world by use of forums, YouTube, an official website, by submitting your game in to the IGF or by plugging in at any indie game expo you can, etc. If your game is good, your target audience will like it and they will support it, if not, they won't.