Is there any chance that a 1300 rated player can beat a 2700 rated player?

Yes, computers can play chess more quickly than a master can.  They can also see deeper than masters can.  Welcome to 1998. :-P

If it's not that hard, then you've clearly done it before, right?

a 6 year old girl from Norway beat GM Simen Agdestein

There is a chance that anyone can beat a 2700 rated player. Lets say someone who does not know the rules of chess plays an infinite number of games against the 2700 rated player. Since he does not know the rules he moves a random piece to a random square on each of his moves. He will win an infinite number of games, because of the theoretical chance of him only doing best moves in a game.

Ok, let's tackle this one at a time:

Greedy, your online standard rating is above average.  You still wouldn't beat a 2700 with pawn odds.  You're not beating a 2700-rated player in a standard control game up a rook for a bishop.  This thread is about someone rated 1300.

letsplaychess: first of all, if your argument includes infinity, you need a better argument.  Second of all, a 1300-rated player KNOWS THE RULES, so they won't make sheerly random moves.  They will use their flawed judgment.  Finally, even if they DID make random moves, the 2700-rated player would make infinite NON-RANDOM moves, meaning that in the end, their judgment would STILL prevail.  Infinity is a nice catch-all that people who don't understand statistics and probability try to use.  

There's a chance that some posters (luigi in another thread) could offer a logical statement backed by evidence, but I would not put money on it. Betting against is too close to a sure thing.

Same with the chances of the 1300.

But if the 2700 blunders his king? Remember , a 1300 is strong enough to survive 40 moves against a 2700, but probably a couple of pawns down with a lost position.

Really? A 1300 player would not make it to move 40 down 2 pawns or less more than 1 in 500 times, at best.

Most 1300s don't last thirty moves in a simul against a 2400.

200 players don't blunder their kings against 1300 players.  It just doesn't happen.  The only example i've ever been able to find was Kramnik vs. Deep Fritz, but even then, that was in a tense position against the best computer out there at the time in a match being followed all around the world.  You're essentially asking, "What if a stop sign was the color nine?"

I'm sure a 2700 player would not want to play a 1300-rated player, because it likely would not be a challenge, so I don't see anyone having a chance to prove any of this.

Also, for the record, I'm rated 1800 on this site in correspondence.  I've had 4 games against an IM rated 2200.  I've been able to survive past move 30 ONCE.  I am 500 points above a 1300-rated player, FearNoEvil is 500 points below 2700, and I couldn't get to 40 moves in four tries.  You're making an awfully large, dumb assumption to just figure a 1300-rated player will make it to 40 moves against a 2700.  The last 1300 I played in G/60 OTB I had crushed in 9 moves, although he dragged it out until I mated him on move 31.  

So which one is it lol -- a 9 move win or a 31 move win.

The 2700 can play because of friendship or money, or it can happen in an open tournament like Politiken cup in Denmark.

The thing about these matchups is that, indeed, the rating difference could make each player play in a completely different way than they normally do. Which I would normally say is part of the game, but here it's like you would change your playing style just to not have to be bored playing the 1300 for too long. I know when I'm playing really casual players, I'm not really in game mode. And that kind of mindset could really affect the game statistics in a way that's just not related to how competitive chess is.

This applies to a much smaller extent to 400 or 500 point differences, but yeah maybe not that much. But when you get really really extreme it could definitely be a big factor.

Still, even if the two players did play like they "normally" would, it still seems weird to say it's impossible for some mate in 1 blunder not to occur anyway -- 2700s do miss things like that in extreme cases. Very very unlikely, but possible.

Eveybody is underestimating the 1300, and thats our advantage. I am 1375 Fide (1755online corr)and I met a Gm once in simultan. He won after I resigned in move ca44.I am planning to build some strenght before Politiken cup and even if I go in at 1375, my strenght shal be pumped up some houndred points (I guess that 1755 online actually is in the neigbourhood of 1700 fide already, but I need to perfect the openings because there is no database to support me otb). If a Gm wants a fast win, he probably gets it, but lots of Gm s play with patience and takes you down slowly. Acually I dont believe the 2700 (Fressinet?)ever will blunder his king, but shit can happen to humans. A 1300 can blunder the king. 

There are reasons for 2700 getting up to 2700, and one of them is probably extreme accuracy, the accuracy of an artist. That accuracy is a very good insurance against blunders. So I am sure the super-Gm will win, but he have to do it. It is not coming for free.

Lowrated players are lowrated for reasons, and one reason might be that they play more lousy games than fantastic ones. But one out of seven games can be very good.

I will prepare as if I should meet GM Fressinet in Politiken. I will read a lot of his games. Like the beautiful victory GM Peter Leko had against him. Of course Fressinet will win if he meet me, but I will give him all I got and will be more prepared than I have ever been for a game.

Take 200 points off your correspondence rating for an accurate estimate of FIDE rating.  You'd be about 1555.  Also, that would make you not a 1300-rated player.  

31 moves until I mated him, after 9 moves he was down a queen for a bishop, hence my comment about having him crushed after 9 moves, but him dragging it out instead of resigning.  Shouldn't have needed that much clarification...