Is solely playing games with long time controls and analyzing them enough?

For example if I dedicated 4 hours everyday playing long chess games and analyzing them, is that alone enough to eventually reach 2000+ ELO one day? No books, no puzzles, no courses. Just long games and analysis for hours daily, week after week, month after month. 

Theoretically it’s possible for sure. 2000 online is not such a high elo. But I think it’s much more efficient and interesting to incorporate puzzles, books, courses.

Idk, I am 2000 online and I feel like my opponents and I play very poorly. Of course, it is an above average rating.

Why not include books in your program, so you can learn from stronger players? Then, when you analyze your games, you will have a store of concepts, plans, and themes that you can apply to your games.

I mean, basically the only way to learn chess is by analysis. When you're reading books or doing puzzles, you're just benefiting from someone else's analysis, so yes, I would say that playing and analyzing is enough. Don't rely on engine analysis, though, because it might help you see some things but it will give you a sense of knowledge without understanding.

I think you have to want it and be tenacious about it. Most of the people I've observed who aren't improving are either, "playing chess for fun" (ie not wanting to improve), or blitz out hundreds of games, doing the same thing every time but expecting better results. So I think no matter what strategy you use to improve, I think having one is the important thing.

Maybe. But your route will be more effective, more complete and probably faster if you add books, exercises and videos. And always think long term, so you don't get frustrated on your journey.
You can do it!

sort of, where are you getting your opening knowledge from? are you completely winging endgames? are you doing tactics puzzles too? devil is in the details.

but playing a lot, especially non blitz games, and analyzing them can get you pretty darn far. most of chess is middlegame, if you collecting all the thousands of tabiyas without relying on chess material, you will get good, maybe not as efficiently but it will manifest.

No.

If you never learn new concepts, you can never apply them in your analysis, and therefore, not in your games. Your playing strength will always be limited by your knowledge.

It's possible to figure those out by solely playing, but that takes a lot of games, and a lot of thinking.

You can learn all the new concepts that you need to improve by playing a wide variety of opponents and analyzing your games. Find out why you lost a game, by studying it, and you'll learn a new concept.

That includes using the best resources available for analysis, of course.

if it was so simple to analyze and learn new things, we wouldn't need instructional materials at all. The problem with this method is that you don't know if what you figured out is correct or not, and you don't know when it's applicable, and when it isn't.

Here's an example for a lot of low rated online players. They sac a bunch of material, and are dead losing, then they mate their opponent because the opponent didn't see a tactic. Then they go "My play must be okay because I confused the opponent and I mated him". No, you were dead lost and your opponent was weak and blundered mate. Then if you try this against a stronger player, they take all your pieces, and then you lose because you don't mate him.

Here's two examples of what that type of play might look like, and how you could draw an incorrect conclusion from the games:

The second game looks like someone who doesn't understand why moves are played in the opening of his choice.

On move 20, why was f4 played? Well, f4 is a common idea in a lot of KID pawn storm positions, and it fixes white's pawn structure so that black can storm later with g5-g4 and open the g-file, for example. But when is it good, when is it bad?

We can check with the engine. It suggests 20...Qh4, 20...Nf4, and 20...fxe4, assessing the three as perfectly equal. Why? Qh4 was not considered because black assumed that white could pry open the c-file and infiltrate quickly. Nf4 was not played because black did not see a follow up afterwards, and again white has the clear plan of infiltrating along the c-file. fxe4 allows the rook to participate in the game, but after fxe4 what's the idea? How can black respond to white's infiltration? Why can he sacrifice the pawns on the queenside- what type of attacks can he launch with his piece arrangements? Secondly, opening the c-file isn't white's only method of attack. He can play c6! as suggested by the computer in many lines. What's the point of c6? Is it to force b6 - but then why is forcing black to play b6 a good decision, how can white exploit this? And how does that measure up to black trying to attack white's king, and does white have time?

You can attempt to use the computer to analyze these lines for you. But you need to be able to understand why these moves are played, you need to figure out when they are good, when they are bad, how to evaluate all of them, and then how to do it quickly. These are not things that can be quickly learned, from just checking with the computer, no matter how deep, because you have not developed a sizeable enough mental database of heuristics to be able to apply them in the situations you get into. Learning from a better player, or watching even instructional videos, these people have already figured these parts out. They can give you ideas of places to look into, they can provide model games to illustrate their examples, a coach can even play positions against you to let you have a feel for how they work.

For example, in a ben finegold lecture about the greek gift sacrifice, he said something that I have since adopted and used with success: if they sac on h7, and you can play Nf6 or Bf5, which defends h7, then the sac usually won't work. If you're only analyzing your games, you might not even see the greek gift (it's somewhat uncommon), let alone be able to figure out a nice heuristic for evaluating those positions. To be able to develop a saying like that, you must first see many greek gift positions, analyze all of them, then notice the commonalities and key squares that make the sacrifice work. From there, you can try various solutions and see when it doesn't work, and what are the commonalities in those positions. If you're only working with your own games and not the games of other players, you cannot possibly accumulate the experience to figure those out. Most people never do in their lifetime, that's why you see people that are sixty, have been playing for 50 years, and are rated 1500.

Going back to the King's Indian game, I don't have a coach anymore. I also don't have a book on the opening. Videos of the opening online do not go into the required depth for them to be helpful to me. If I Want model games, I can search in the explorer - already breaking the premise of the original post - but even then it would be difficult for me to find good model games. Some games might be decided by silly blunders, some might be decided by exceptions to the rule, and lead me down to a wrong path. And if that happens, how will I possibly know, if I don't have outside input? The only way for me to figure out if my methods are wrong, is if I lose enough games that it no longer seems to be by chance. But if I keep winning by fluke tactics as in the game posted, would it even be possible to figure that out? It's very easy to say that your play is "good enough" and that you want to "work on more important things". Then you never learn more about the positions.

How many games would I have to play through, that use the f4 push? How many will I find where the f4 push is bad? Probably not that many. And even if I did find good games, I'd still have to find a way to explain to myself how I can evaluate these kingside-queenside pawn race positions. Very difficult to do!

It's a lot better to learn things. It will make analysis a lot easier. Let's say I figured out when f4 is bad, because of X, and f4 is good because of Y. I might notice in a game I play in the future that I missed an opportunity to push f4, because the position had X characteristic. Then, I can be more alert to X characteristics in my King's Indian games. That's when analysis is at its most helpful. But it's difficult to even get to the first step if you're not learning from external sources.