what's the main difference between a 1300 and a 1800 player?

All chessplayers have variable strenght. I am on average close to 1300 Fide, but in my best game I am closer to 2000, and in my worst games below 1000.

A 1800 is more consistent I guess, and maybe he is dipping between 2300 and 1200.

When a 1300 plays at 1800 strenght and the 1800 plays at 1600 strenght the 1300 player can win.

This autumn I have beaten one 1800 player in OTB longchess. He played well and I played fantastic. My opening was so spot on that I was floating the first 20 moves. After that it was hard work, and it was a mate in 78.

Well done! Don't look down!

I feel that I have good chances against far better players  if I dont loose an officer in the first 15 moves.

Higher rated players think just as much about squares as they do pieces. That's a huge difference I have noticed besides the more obvious things like less blundering, better at development, more opening and endgame knowledge, etc.

Thinking about squares and which pieces and pawns control what squares is something weaker players think much less about. 

Anyone can spot tactics given enough time, they'll see the fork or double attack that can be made. But getting to those positions requires good development, positional play, and a strategy.

Most lower rated players don't know what to do after they have developed their minor pieces and castled. They don't know that they should also develop their rooks to central squares, and usually move their queen up a square to connect the rooks.

They trade minor pieces (knights and bishops) without thinking of the pieces strength, or with a strategy in mind. For example if you have a bishop that is closed in, and your opponent has knight on the 4th or 5th rank, the knight is more powerful than on the 3rd rank, and if your bishop doesn't have much room to move, it might be a good idea to take the strong knight with your weak bishop.

Or in another case maybe you have a strategy to capture the enemy's kingside knight with your bishop to weaken his kingside. Those are two examples of good trades and things that a lot of weaker players do not think about, they'll just trade pieces "willy nilly" just to make moves.

In the endgame the biggest beginner mistake is not bringing the king into the action soon enough. If I make 2 or 3 moves with my king before someone makes their first king move, that will likely give me the advantage. I'll have better chances at capturing enemy pawns and promotong my own pawns.

WIth all due respect, I have a hard time imagining the flux being more than 300 points in either direction.

At my level, I am my worst enemy, and one blunder do cost more than 300 point strenght. If I play clean, good and blunderfree, I usually beat players rated 300 above me(otb-rating-not chessonline, chessonline I have problems with beating +200 even if I dont blunder).

Today I am online 1670. I am playing a game now against a 1420, He did blunder (fork). I will win. A couple of games ago he didnt blunder, and he won. He is very close to me in strenght when not blundering. He is absolutely capabel of playing at chessonline 1720+ if he dont blunder.

I did play a 20 min 5 sec tournament this autumn, I listed one class up (B, not C where I belong), and had a ratingperformance of 1702. My Norwegian rating is 878, which is a strenght that is probably below 1400 fide.I am not fiderated yet.

http://ratings.fide.com/card.phtml?event=1524283

They have an expected score of 8%.

So 8 wins in 100 games, or 4 wins 8 draws.

However, the rating system doesn't work well with large gaps and above about a 400 gap the higher rated player will score more than they should.

Lower.

Typically the rating on chess.com are much more accurate than Otb-ratings like Fide, UCSF, Elo, because the internetratings are fresher, and continously updated, and is typically based on a much larger number of games.

In other words, I am better than I am in real life.

Hello I'm a Mahmut wher are you from?

main difference is that one is way smarter

Which is why he isn't 1700.  He DOES blunder.

Yes, but only in the games he doesnt win. So his actual strenght does vary very much from game to game. 1700-players (chess-online)are more consistent, but not neccesarily better every day.

So if you take away all the games he loses, his rating would be higher.  Well, of course.    

My own experience, and evaluation of players I met here, otb, and in my otb-club.

On chess.com the rating is updated the second the game is finished. That is a fact. Fiderating is updated far less frequent.

The huger number of games , the more accurate the statistics. That is a fact. My nephew competed active in chess for several years, maybe 8. He got ca 250 matches otb, and his rating are based on them.

This year I will with less effort end up with more than 600 chess-onlinegames.There is no way I could played more than 200 otb-longchessmathes in a year.

My Norwegian rating (878) are based on 12 games, during the first half of 2014. It does not reflect my current strenght, which probably is above 1100 Norwegian, and I expect this new rating to be official maybe around newyear, several months after I actually got that strenght.

The main difference is the timedelay. Fiderating is accurate for players who has the same strenght as they had last year, and that goes for a lot of the old guys.

The new guys, freshly downloaded from the internet, and the kids, the ones that has full speed forward are months ahead of their fiderating.

The statistical term you are looking for is reliable.

Nonetheless, the reliability extends only to predicting results within the rating pool, not across other pools.

The main difference is 500 points....any other questions ?

Hope this helps.

Ed_Seed, your comments are on the mark, but Djonni is correct that a large number of games is relevant. My central point: ratings across pools cannot be compared. The rating in one pool cannot be more or less accurate than that in another. Nonetheless, a rating based on a higher number of games offers a better base for predicting future results. Glicko rating systems are preferable to Elo because they take level of activity into account.

The point Djonni was struggling to make concerns not size of pool, but number of games. Aside from that, we agree.