If Carlsen is as strong as his rating says and is as far above his peers as Bobby Fischer was to his then I would say it is impossible. Does anyone know the last time Carlsen lost to an International Master? No 2000 player could beat Carlsen in a blindfold simul much less in a tournament. Period.
Could a 2000 rated player beat Magnus Carlsen?

No 2000 player could beat Carlsen in a blindfold simul much less in a tournament. Period.
http://www.chess.com/blog/Gilmar8Andersen/magnus-carlsen-lost-to-a-beginner-in-brasil-in-a-simul
well,if you count a 2000 rated player a beginner...well im a retard lol
Not only could a 2000 rated play not beat Carlsen but a 2200 player wouldn't stand a chance if he is as good as advertised.
Question (Reddit) - How many beers would you need to drink for a Master to be able to beat you?
Carlsen - "There is no answer to this question... even extremely intoxicated my chess strength and knowledge is still in my bones."
http://www.thechessworld.com/learn-chess/18-general-information/362-33-questions-answered-by-magnus-carlsen-on-reddit
An interval or rational scale measure the same thing over time. A fever of 103F doesn't have an inflationary value where that becomes a normal temperature 20 years later.
Even if one granted a normal distribution it is still obvious that the _qualitative_ difference between a 1000 and 1800 and 2000 and 2800 is different. An 1800 is going to have some knowledge of openings, tactics and perhaps end games. A 1000 is probably going to know how to make the moves. When An 1800 is playing out a number of standard moves from book she is essentialy playing out grandmaster and computer analysis and thus playing well beyond her strength in that part of the game. When we get "out of book" we fall back on our own resources.
The _qualitative_ difference in the game between a 2000 and 2800 is likely going to show the 2000 playing relatively strong moves in certain parts of the game while playing like a duffer in others. A 1000 rated player is going to suck in all aspects of the game-- a qualitative difference.
Claiming a standard distribution doesn't make it so. Modifying equations to artificially create a standard distribution from current scores doesn't make that a true equation. It has descriptive power but predictive power is far more limited.
I've been reading every update to this thread since I made the mistake of writing in it and this is one of the only posts that even comes close to understanding the situation.

Well Daniel Lampert, then-rated 641, beat Nakamura. Hikaru has also notably struggled against 1300's, scoring just over 64%, while performing slightly better against 2000 players, at 69%. So yes, a 2000 could beat Magnus Carlsen! (In the 20th century perhaps)
http://www.uschess.org/datapage/gamestats.php?memid=12641216

Daniel Nordquelle (2049) has what it takes to win against Magnus if Magnus blunders as bad as he did against Switzerland yesterday.
Yes, a 2000 player could beat Carlsen, and it's probably already happened.
Oh, you mean in the year 2015?
(Or @ the OP, in the year 2013)
i am rated 1500 and some thing in this server ...i dont play alot of chess i rather solve puzzle and watch games
most of the time i play chess against computer
i really get some wins vs computers that are 2000 i am finding great difficulty beating humans at rated 2000 but let consider this failure due to my continuous play against pc but i develop this patten
against 2000 pc is beat him 1 "majestic chess " match each 10 games
against 2200 i beat him 1 match"kasparov chess mate" each 12 games
against 1900rated program "like chess free app on android which" i tbeat him 3 to 4 games each 10 games
against 2600 i lost all games i only draw 2 games vs chessmaster 10000
so above 700 difference in rateing your chances of wining alomost dissappear
way far than 700 point of differnece your hope is just one percent
and that percent is luck ...which hardly happen vs pc
but any way carlsen is human and mistakes does happen and
remember that carslen is more speacilaized beating high
rated players who make strtegic menuvering like carseln
but lower players rely more on tactics which may be more bothering to
carlsen
breif ...dont hope much...but any thing is possible
and here we are in it
jestus williams 2300 have beaten magnus carlsen ..the difference
is just 5500 points which again in my thinking
as long as the difference between 2 players is less than 700 hundered
you get a chance to win a game in 10 ...justes was lucky to pick his
win so early he lost the first 2 blitz games and then win the third
also we must know that magnus power lies in understanding the style
of the high rated player those who make to many startigic manuvering
more than those who rely on tactics ...in magnus own words
"i play online and i woudnt tell you against whome i lose"
clearly he refered to lower rated players 2000 and higher
If Carlsen had skipped his morning cofee and underestimated his opponent, which the 2000-rated fellow would tactically utilize, then absolutely.