Do the best game developers constantly follow feedback and listen to what the fans want, or do they just follow their own ideas and have a dogmatic stubbornness when it comes to following other's suggestions?

Wow, you asked that question in a pre-decidedly lopsided way. Do they <insert positive words like "follow", "listen", "what [people] want">, or are they evil nazi baby-murdering jerks?

What exactly makes it "dogmatic stubbornness" that they don't listen to ideas? What makes your view on the matter anything except dogmatic and stubborn for being predisposed and biased against developer's perceived choices?

Maybe you've already made up your mind, and are either venting or soapboxing. But if you are actually seeking alternative viewpoints...

A) Are you sure it's the developers, and not the managers who tell the developers what to work on? Or perhaps it's the lawyers who advise, for legal reasons, not to read the forums? (Not sure if that is something that actually occurs or not)

B) Are you sure they aren't actually listening to ideas, but just aren't responding to them? Maybe it takes them three or four months to actually implement them, and meanwhile people are now shouting for something else. "It's about time you implemented that! Took you guys long enough. Have you read my latest list of feature demands yet?"

Or maybe the developers are so busy implementing the hundreds of ideas they have, that they don't have time to get around to the millions of ideas everyone else in their armchairs are having?

I mean, I get loads of ideas all the time - I'm sure most players AND developers do. But it takes time to implement each one of them. Far more time to implement then to think up. A continual flow of ideas, with each idea taking longer to make then to imagine - that guarantees every idea can't be implemented.

C) In studios where the developers actually have creative control, why shouldn't they work on what they themselves want to make and play? Are they actually morally or legally obligated to make what you want to play? Or what the mob wants to play? (Ugh, not another twenty dozen Angry Birds, Call of Duty, and World of Warcraft clones)

D) Who's to say the vast majority of gamers aren't already satisfied? The public and developers are separated by a wall between two different rooms, and the developers hear shouting on the other side of the wall, "Do this! Remove that! I hate this! I loved that other thing! Jeesh, you guys are ruining the game!". But whenever they actually walk into the other room to get a better idea of things, the find that 90% (statistics made up) of the gamers are sitting down quietly and enjoying the game, and only the 10% who aren't liking it are shouting their complaints. If consumers shut up when they enjoy something, and shout when they don't like something, and you hear a bunch of shouting, all the shouting will always appear predominantly negative... because the satisfied ones are quiet. Regardless of how much negativity you hear, that doesn't tell you at all how many satisfied players there are. It might even negatively re-inforce the shouting, because 'Of all the people talking, everyone agrees that this is bad, therefore we must be right!', but the only ones talking are the vocal minority that don't like it, not the fully satisfied majority that are too busy playing the game to bother informing the developers, "We like the game as it currently is."

E) But, alright, let's suppose for a moment that the people shouting for changes are actually in the majority (which usually, they aren't, despite their own views that "EVERYONE" agrees with them). So, let's listen to their shouting for a moment. Pretend you are a developer with a popular game with a very large and established community. Now, pretend you are standing barefoot in the middle of a large room with shattered glass all over the floor. The shattered glass represented game mechanics that wouldn't work well with the game you are running. Now, there are multiple safe paths through the glass to multiple great game designs that'd work very well. The only problem is you are blindfolded. You can trust in your own memory of the paths, but you're memory isn't too great, and you can trust in your own experiences of where you step and get hurt and know not to step in the future.

Or, you can, blindfolded, listen to the crowd around you. If you listen very carefully, you begin to hear them give you directions: everyone is shouting different directions. 15% are shouting "Go NORTH! Go NORTH!", 20% are screaming that "SOUTH IS BETTER!", 15% are saying "ANYWHERE EXCEPT WEST!", 10% are saying "EAST OR WE'LL QUIT AND NEVER COME BACK!" (but they are deluding to themselves, and will play regardless), 25% are shouting "LEARN HOW TO FLY!", 10% are yelling something about rhinos and chipmunks, and the final 15% are saying, "I don't know what you should do, but I hate where you are currently standing.".

It's the developer's bloody feet (costs of development), and wise developers will tune out the chaos and take very careful steps, cutting their feet on bad ideas (most of which player's will never know were already tried and rejected before even being made public), with the developers gaining experience, and coming to their own informed decisions - decisions that, while maybe not being perfect, are still good decisions that please the greater majority of players... with alot less pain.

F) But alright, let's suppose the players really are unified 100% on what they want. The developers still shouldn't listen to them.

Why? Because if you ask players what they want, they'll give you a laundry list of last year's gimmick features. (Sidenote: If you ask a manager what they want, they'll give you a laundry list of three years ago's features... and then go make you work on it despite your objections) (Note 2: I don't actually work in the industry, I just wanted to get in a few jabs at managers, since it's all the rage nowadays).

Gamers want what is already familiar to them. "Make it more like Game X! I enjoyed that!", "I like that thing from that one game I played when I was younger. Man, they sure don't make games like they used to!", "I think you should do FeatureY, it really worked in GameZ!"

For games to actually grow and improve, there needs to be room for people to try new things, not just continually re-hash old features - even if those old features were popular.

G) Game features work together as a cohesive whole. You can't just take a bunch of features, that may indeed have worked well in other games, and cram them together with ducktape. The game will be unbalanced (even if it's just a single-player game) and convoluted, and ultimately unenjoyable. You need someone who can look at the whole picture and make unpopular decisions to make the game overall better.

H) Many people don't know what they really truly want, and think if they try this or try that then it'll suddenly fulfill their longings. But not everyone who tries to decorate their house are actually good designers (very few are), and often they just clutter their house with a mishmash of conflicting ideas because they don't actually know what they are doing. (Even worse is when they find out they enjoy ruining the decor of their house, and then think that because they enjoy it, they must be good at it!)

Most people think they know what they want (and are even very passionate about it), but don't actually have a clue. If a player says they want something, take it with a very large grain of salt - they are guessing that it might improve things for them. But if a player says they hate something, listen carefully - because they know that what is currently there is not satisfying them.

Listen to complaints, and get great detail about what's not working. But, for the sake of your bloody feet, be very careful about what conflicting ideas people are shouting based off of pure guesswork they have that they might like it if you sortof make it a hodgepodge of what they used to play in other games. Especially be careful, because it might disrupt and annoy the 90% of your fanbase who are already satisfied and playing quietly.

Ignore is perhaps too strong of a word, but at the very least, don't just adopt their "fixes" and "ideas" wholesale. But do accept their facts about the current failures (but bear in mind that even their complaints might just be personal taste that the silent majority enjoy).

...or maybe developers are just being dogmatically stubborn, because they are just evil greedy jerks and not capable of intelligent thought, and make their decisions purely to try and annoy you.

Thats a very interesting question..

You can do it from a market point of view, where getting money is the goal, or at least needed, and so, you NEED to make the greater number of ppl possible be happy with your game. Doesnt mean blindly listen to feedback thou, ppl are stupid, they dont know what they want, they just know what they DONT like AFTER trying it. Of course, theres hype too, and its a powerful technique, where you can play with the fact the ppl dont know what they want. They rarely know what they do like, even after trying.

Or you can do it to please yourself, in that case your opinion may not be friend with the majorities opinion, and you just want to pull out that idea you have. Hoping that enough ppl will share your feelings for the game (so it may became profitable). Its important to not lie to yourself here, saying you like your extremely shit game..External views may help open our eyes for issues you didnt realize.

Game design is very organic, and evaluation of design is very, very fuzzy. When you're doing your initial pre-planning, you're constantly going to be coming up with wouldn't it be cool if's. You have to continue on at some point, and lock some things down. You also have to be able to throw out terrible ideas, even if literally everyone including yourself thinks they're great (again, "fuzzy") at the moment.

You have a budget (yes, even a one-man indie operation has a "budget" of some kind) that you have to keep to, likely someone you have to answer to, and are frequently not in total control of your own project, creatively or technically. Even a one-man operation might have the looming threat of angry kickstarter backers.

Given this budget and a timetable, you have limited time to get everything implemented that may or may not be a good idea, and new ideas are still going to be coming to you. Some of those ideas you might actually benefit from implementing, but many are simply too expensive to do, as features tend to get more and more expensive to add the closer you are to completion.

Then, your game comes out, and suddenly ten thousand people are suggesting conflicting things on what would make your game "better". Turns out, making games is hard (who'd have thunk it), and even the best devs make mistakes. In fact, it's physically impossible not to do so, since evaluating "fun" is going to change day-by-day ("fuzzy"). gamedev isn't some precise plan like Operation Overlord, it's a huge mess of plans and kneejerk reactions. It's freaking World War 2.

Gamedev is art, insofar as it's the practice of taking something mechanically impossible ("fuzzy") and turning it into something that "turns out pretty good".

As for me, I don't put forth a lot of creative input (or even work directly much on the games themselves) professionally, but in hobby work I put myself above all else. If I don't find it fun, I'm not going to put it in, at least not as a mandatory mechanic. Even if I went indie and tried to commercialize my stuff, I work for the love of the craft and product, not as a factory for making spreadsheet-processed lowest-common denominator hotdog fun.