I've heard it before, that the difference between 2200 and 2500 might be just 2 or 3 key positions in a game where the 2500 can find the best move and the 2200 can't.
Looking at chess this way it's a bit discouraging that such a thin line separates players. You might have to work 5 years and/or on a few thousand more positions to squeak out the extra 2% of accuracy just to get to the next class.
Consistency is the key. Anyone rated A class and above knows what they are doing. So what separates the A class from the expert? The expert does less blunders,and less innacuracies. But I bet you they both have the same understanding of ideas in key positions.
And yes,the gap from 2800 to 2000 is huge.
"Inaccuracy" is just an euphemistic way of saying that it is a sign that the player is running out of ideas...? This usually happens when the position is already slightly worse anyway, and the possibilities for ambitious play are slowly decreasing
^ no, that's not what it means. Please don't give words your own meanings.
Well it has been my own observation that what people call "inaccuracies" are the (logical) consequence of a position already sliding away from the hands' grip. Being a chess player, I am aware my observations may well be based on mirages and that my tentative conclusions may have been made without enough consideration
I've seen, a lot of times, games where a player, quite literally, 500 or more points lower than their opponent, reaches an even, simplified position, but, sooner or later, goes on to lose. The point is, if you are constantly being set with problems, even if you have solved the first dozen of them, eventually you will crack to someone who is better than you. And of course, you can play an entire game well and mess it all up with one misunderstanding or two. But that's ok -- I don't mind a challenge.
That 1500 playing the 2100 from that equal position (something I actually witnessed) -- blundered a piece or something eventually. He played just as well as the 2100 for the first 50 moves, but in a couple of moves he lost it all. And I knew that was going to happen. The same 2100 got into a similar situation against a 1600 (and me as well, but I would prefer to remain silent on that one... :D)
I was in a pretty drawn position against a 2300 (I was slightly worse but the position was too simplified for him to be winning) -- just two subtle endgame mistakes lost it all for me. But it was kind of inevitable. In the same tournament, same round, I saw something even more unbelievable than my experience -- a 1900 had what seemed to be just a slightly worse endgame position (material equal) against a, literally, 2650 USCF player. But I knew that grandmaster would grind as long as he needed to, to eventually claim the point. Ratings are based on results; they don't ask how you get them.
I think there are certain parts of a chess game that highlight your strengths, but the weaknesses drag you down. I was doing very well against the 2300 until the game got into my weakness -- the endgame. Before that, a man sitting next to me couldn't tell who was the master! (he told me some time after the game) But the key is "before that"...
I've seen, a lot of times, games where a player, quite literally, 500 or more points lower than their opponent, reaches an even, simplified position, but, sooner or later, goes on to lose. The point is, if you are constantly being set with problems, even if you have solved the first dozen of them, eventually you will crack to someone who is better than you. And of course, you can play an entire game well and mess it all up with one misunderstanding or two. But that's ok -- I don't mind a challenge.
That 1500 playing the 2100 from that equal position (something I actually witnessed) -- blundered a piece or something eventually. He played just as well as the 2100 for the first 50 moves, but in a couple of moves he lost it all. And I knew that was going to happen. The same 2100 got into a similar situation against a 1600 (and me as well, but I would prefer to remain silent on that one... :D)
I was in a pretty drawn position against a 2300 (I was slightly worse but the position was too simplified for him to be winning) -- just two subtle endgame mistakes lost it all for me. But it was kind of inevitable. In the same tournament, same round, I saw something even more unbelievable than my experience -- a 1900 had what seemed to be just a slightly worse endgame position (material equal) against a, literally, 2650 USCF player. But I knew that grandmaster would grind as long as he needed to, to eventually claim the point. Ratings are based on results; they don't ask how you get them.
I think there are certain parts of a chess game that highlight your strengths, but the weaknesses drag you down. I was doing very well against the 2300 until the game got into my weakness -- the endgame. Before that, a man sitting next to me couldn't tell who was the master! (he told me some time after the game) But the key is "before that"...
Interesting point.
I know what you mean about getting good position though :) If I travel to a tourney with someone, I like to tell them about my loss vs a higher rated player by starting: "well I had a great position until I started losing" 
I find this slightly amusing. The differences are huge. I recently read a book where the author describes the gap between a scratch golfer (0 handicap) and a pro. Very detailed and revealing.
In any game or sport, the "curve" becomes nearly vertical among those that are good, compared to the rest.
Obviously not all my games are like this, but still improvement from 2000 and up must take incredible consistency
Indeed ! 
It's not easy to throw stronger players off-balance... But consistency is key, if only because the FIDE rating system is merciless when it comes to losing against weaker players.
Consistently drawing inferior positions against lower-rated players is the only way to move up, so yes, consistency is needed...
Waffle is that game available to view on your profile? If so what's the game, I'd like to take a look if possible.
I didn't save it directly, but I know it put it into my cpu to look at, so by using the date of my OP I may be able to dig it up. Not at home right now, and it would take awhile. If I come across it I"ll post it.
Part of the point I wanted to make with my examples of lopsided encounters was that just because I was "even" with the master so long, doesn't mean his 400 rating point edge over me wasn't legitimate. I slowly, but, in a sense, rather inevitably, lost. And indeed, it had to do with consistency.
Me bragging about something like that is like a beginner being proud of "surviving" for 50 moves against a "strong player" -- the beginner could be down a rook in an ending, but sure, if he plays until mate, then he can hold out actual checkmate for some time; but at that point there is no question of what the result will be, sooner or later.
Remaining consistent during games can be difficult, especially for beginners. I suffer a lot from this during some of my games.
Some moves can be blatantly obvious but often they are not. Usually you just want to improve your position whilst stopping any of your opponents immediate plans. When the position remains abstract it becomes increasingly difficult to find the right move, as a result it's usually the weaker player that will snap first.
The stronger player will maintain the psychological tension of the position and adjust their pieces accordingly. Often we think we have to run at the enemy king, we have to "make something happen", run a pawn, move the queen in to attack. But often this is actually helping our opponents.
When the position explodes and becomes concrete that's when we should look for more tactical lines, but still remain consistent, and be patient. You might have the right idea and even the correct sequence of moves, but it doesn't mean your going to play them in the right order.
Think about the order in which you will re-capure, does it leave holes in your defence? Does it lead to progressive tactics or checks? Is winning the pawn really worth compromising the position?
Remaining consistent is what separates the experts from the rest, don't be frustrated with it. It's tough losing a game when you know you have played strongly, solidly, to an opponent who was just slightly more accurate/consistent. However if we learn from those games, we can almost always see a bunch of moves that would have changed the face of the game.
Part of the point I wanted to make with my examples of lopsided encounters was that just because I was "even" with the master so long, doesn't mean his 400 rating point edge over me wasn't legitimate. I slowly, but, in a sense, rather inevitably, lost. And indeed, it had to do with consistency.
Me bragging about something like that is like a beginner being proud of "surviving" for 50 moves against a "strong player" -- the beginner could be down a rook in an ending, but sure, if he plays until mate, then he can hold out actual checkmate for some time; but at that point there is no question of what the result will be, sooner or later.
Heh, exactly. That's why I like to joke "I was doing well until I was doing poorly."
I can get equal, and even superior positions against players better than me all the time... but there's still a lot of chess to play, and as Tartakower said, the mistakes are all there waiting to be made!
This is what I was starting to expect. Almost everyone can hit a bullseye its about how often you can hit it.
I find it slightly amusing that alczervik the weaker player out of himself, elubas, and waffle, thinks that reading a book gives him more insight into this idea then the players who are closer to master strength.
al, would you trust a golf pro about if golf was about consistancy or an author with a few theories?
This is what I was starting to expect. Almost everyone can hit a bullseye its about how often you can hit it.
I find it slightly amusing that alczervik the weaker player out of himself, elubas, and waffle, thinks that reading a book gives him more insight into this idea then the players who are closer to master strength.
al, would you trust a golf pro about if golf was about consistancy or an author with a few theories?
Good point! Practice makes perfect!
http://enjoychesslearning.wordpress.com/
Played a pretty strong player G/20 online yesterday. Saved the game and reviewed it with my computer. Now sometimes the difference between computer choice #1, #2, and #3 were significant, but I was amazed that almost all my moves were in the top 3 yet I was still ground down and lost. (I enjoyed the game, this guy was really good).
Also, in the game the ideas of the middlegame and endgame were pretty clear to me, and of course when you're following the thread so to speak it's not hard to avoid blunders.
But I'm easily below expert level... and there is really that much difference between 2000 and 2800?! It's amazing. Obviously not all my games are like this, but still improvement from 2000 and up must take incredible consistency. Top amateurs (and those below them, players like me), who are considerably worse than say, IMs, are actually a lot better than I gave them credit.