I've been working on my game since the Release 1.0 came out earlier this year and so far have been focused primarily on the gameplay.  Just recently (last week) I started adding "polish" to various areas of my game.  I'm getting a little worried that my game does not have enough graphics bells and whistles at this point (bloom effects, etc.)

I'm curious to know how the rest of you are approaching your XNA game dev projects.  Are you focusing on graphics (eye-candy) first, gameplay first, or a mix of the two, or maybe even some other order of events?

GAMEPLAY!  ;)

Of course, everyone will say Gameplay over Graphics :P But in any case, I'll have to go with everyone else and say Gameplay first, proceeded by awesome graphics.

Another vote for gameplay here. You can get away with great gameplay and crappy graphics. You won't get away with the other...

If you don't finish your project, a game without good graphics is an ugly game.  But if you start with graphics first and not finish, you have a nice slide show... ;)

DOAX2 anyone? While not a terrible game, it's selling point is obviously the pleasing polygons, not so much the gameplay ;)

I've never played that game, but I've been hearing evil things about it :|

I feel the need to play devil's advocate here and say graphics.  I start out with really basic made in paint/blender graphics and then work on gameplay.  Cause you don't need a dazzling level if the mechanics are whack.  On the other hand you can't have killer gameplay if you don't having something to show you visually that the code is performing up to snuff.

Even Gears of War sandboxed levels on the fly to test the mechanics before actually fleshing them out to their finished states.

I do agree that killer graphics can always come later, that's what the artist is there to do.

RinusMaximus:

Another vote for gameplay here. You can get away with great gameplay and crappy graphics. You won't get away with the other...

QFT!

I'll take a slightly different approach here and say, "focus on whatever strengths your team has first".  And here's what I mean by that.  If you're a lone coder ... then chances are that you'll definitely want to focus on gameplay first while using some janky programmer art. 

However, if you're lucky enough to be working with a team (or at least you and an Artist), then you should focus on first delivering one solid piece of the game.  That means menus, awesome art, 1 good level (or whatever for your gametype), and solid gameplay *in that level*.  The idea here is that you get one bitesized snapshot of what your game could/should be, and tweak from there.

If you're curious about this methodology, check out http://agilegamedevelopment.com/

When considering this question, you have to ask yourself, of all the games you own, are you willing to pick up and play an older one, whose graphics are dated?  If the answer is yes (and for 99% of us it will be) the reason is because of the gameplay.  So, keeping that in mind, gameplay comes first.  If you go to extremes and make a game that's 100% graphics then it's not fun to play, but 100% gameplay is, even if it leaves you wishing it looked a little better.

I'm a team of one and now that I have some basic gameplay in place, I'm bouncing back and forth between gameplay and graphics. Mostly I'm doing this to give myself a break from one task by focusing on the other. I'm no artist, so I find the graphics quite challenging. If I left all of the graphics until the end, it would feel like a huge, discouraging task.

yeah, ultimately everyone kind of works their own way.  When you're by yourself, really, whatever feels right is the way to go.  The challenge comes when you try to assemble a team ... now one must deal with a number of differing worker types.  Actually, Vic from the FlatRedBall engine (XNA version coming soon!) posted a pretty good article on how to successfully lead an independent game team:
http://www.flatredball.com/frb/docs/index.php?title=Tips_for_Completing_an_Independent_Project

I will have agree with MagWheels. I'm also doing a game by my self and I find that it's more productive to bounce around both, I usually do a week of gameplay a week of content. That way I can get a greater perspective of the whole game and mix the gameplay and graphics in an better way. I believe that if you are working on a problem for more than 1 hour you should do something else and go back to it latter, that makes me gain time and don't be all stressed out about a stupid error.