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

The point is 1300 wont make random moves, he will make random moves from the set of the moves he can see. %90 of the time best move wont be in his horizon, %20-30 of the time he even wont find a plausible move. In other words, he can find %70 of the playable moves with great luck, or even find %10 of the best moves with great luck but %100 of the time he will play bad moves at least %20 of the game even if he is superlucky, because he even wont consider normal enough moves.

I dont believe 1300 can win, but hypothetically if 2700 blunders more than and 1300 can blunder with 10 blunders in 1 move it can be. Like İf he gets hanging queen with check and captures rook with already some pieces are exchanged, even if 1300 blunders a few minors and pawns he might win etc.

Super lucky 1300 vs Worst ever blunder 2700 might be a struggle. ( Still not sure if 1300 can)

By the way i just seen a 1400 slatemated against 1100 a queen up (and only queen on board) yesterday right after he queens. And we are talking about 1300s even that position doesnt ensure he can win %100 :D

When i played Fabiano otb he blundered a knight and I would have won had i not blundered. He is over 1000 points above me.

Speaking scientifically, if a patzer 1300 played 50,000,000,000 games against 2700 rated players, he would win a few. In the grand scheme of things, that is actually a low number (50 billion). 50,000,000,000,000,000 games and he will definitely have one or two winning chances.

So, probably more rare than winning the lottery but more common than 0%

50 billion games may bore the strong GM to death, but without such a health risk, that's too few to bring about a reasonable chance for a draw for the 1300. A win remains far beyond the horizon.

In person, with dirty tricks, maybe a very, very slight chance.

I'd go farther than that and say it's much more likely than most people probably think it would be.

the only chance the 1300 player has is if he's cheating, or sandbagging. or the 2300 player dies.

1300 player can and will beat a 2700 player if the 2700 player doesn't think and just pushes wood while underestimating the 1300 player because of his 1300 rating

A bit over a year ago. I mean, obviously there is a lot of speculation you have to do, but I would say for humans it's pretty darn easy to forget how big even numbers in the thousands are.

My game was an easy win. But it doesn't make a whole lot of sense to put so much stock into one game -- it's not at all representative of what thousands of games is like. I think there is the illusion where we think that if a task is easy, then doing it thousands of times successfully, in a row, is easy. Yet it's probably not. Not when all of the liability is on you, when any little weird problem that comes up could ruin your entire streak. The journey to thousands of wins in a row (which could run over the course of years or whatever, I'm not saying a thousand wins in one sitting or anything), even against weak opposition, it'll seem almost endless. It'll seem like every win you get is meaningless because you have so many more games to go. And every now and then you'll come across a small problem here and there, and I think eventually you would start to get scared, start to wonder if a true disaster will end up happening, and without a clear end in sight, as if a disaster is just inevitably waiting for its moment to strike... well, it's just not so easy.

Thousands of games without a loss is the sort of challenge that could plausibly do that to a person. A few thousand is a rather huge number, especially as regards chess games. I might be wrong and maybe a few thousand wins in a row against a 1300 really isn't so bad for a 2700 player, but when dealing with numbers that are so hard to visualize, my conclusion isn't going to be so clear.

I'm not trying to convince myself of anything; I don't care what the answer is. It would be cool if I could just pretend that I had any decent grasp of what a few thousand (non blitz) games are like, but without any sort of basis, I find that hard to do.

So basically it goes both ways. You can say, hey, the chances of me losing seem impossible, so I can't imagine how a loss could ever happen. Or you could look at it from the perspective of how many games in a row you have to win -- it'll be so large in amount that it nearly resembles infinity, and with infinity, well, we know everything eventually happens there.

Now, a large number of games is not actually infinity. And a merely large probability of winning is not actually a 100% chance of winning :)

Just so you know, this post made my day. /thread.

I'd like to recount that on an old account that I later deleted, I opted to play keyton(I think that is his name), a strong player rated 2000. Now I am very easily 2000 at my highest level of play(I'm working on getting my chess.com rating there to prove it, I'll have it done by the end of the month.) But keyton didn't know this, he thought I was a 1300. In all honestly, when he crushed my prized Benko Gambit the first game, I was in shambles. I must have played like it. Lost 2 more games right out of the opening because I was so destroyed. The next game was going the same way, I had an awful looking position and had lost my wits(It actually wasn't that bad, but I was tilting like a jenga tower)...God intervened. I won the fourth game "by mistake". Keyton blundered on the spot, overconfident from so many wins. I took his loose knight and sighed a sigh of relief. I had accomplished what I had set out to do. Keyton probably suppressed the game in question. He told me I had "gotten lucky" and that I had "proved nothing". But it just goes to show...

Let's see.  Imagine the 1300 level player is badly underrated, and that the 2700 player is playing a blind fold simul and he is high on crack...

Supposedly my USCF rating is 1337. I haven't touched or see a real life chess board in 6 years!!!

"And in any case infinity makes all sorts of problems if you try to use it in statistics."

My example of infinity was just to provide some sort of intuition. If you had to do an easy thing a million times in a row, and the chance of you failing once was one in a million, you might succeed, but you plausibly might not as well. But, probably because it's easier to perceive easiness than it is large numbers, we mistake a 1 in a million chance of failing a task with "this task is so easy a fail isn't possible." And sure such a mindset tends to work fine, except in the case of this example, where the task has to be performed 1 million times. When I say a million tries is "like" infinite tries this is analogous to saying a near zero chance of failing is "like" zero. But we see that, in the case of this example, you need to look at both the low likelihood of a fail, and the large amount of times you have to succeed, or else we will probably overestimate our chances of success.

So I guess one good reason why I say we underestimate the 1300s chances against the 2700 is because of that sort of fallacy I just described above. We tend to weight likelihood over large numbers, if that makes sense. If we put more weight into large numbers, we would be more open minded in our estimates than we currently are.

"And infinite games only matter if the person plays randomly thereby making all combinations possible. Humans don't play like this"

Well in some sense they do, actually. We have certain procedures we do in our heads -- for example we may check for all checks and captures, or at least tell ourself to do this. Yet we still miss checks and captures. That is pretty typically human chess, actually!

I play a hundred games a year against players rated 1000 below me. I only fail to win while I'm not attentive to the game.

The OP for this thread describes somewhat different conditions than me chess classes.

"but easily in the hundreds."

But not in the thousands.

It's true! I deviated quite heavily from the original question merely in response to Elubas. It was an interesting anecdote and I intended to supplement his point with a real world example and I think it worked. You are right, world class GM is differently than chess.com 2000, but I think this questions leads to another, more sinister: what are ratings actually measuring? We learn from Andrew Soltis that a master does not calculate more lines, know more theory, etc...than, say, an 1800. Masters have acquired a knowhow that allows them, in essence, to win more. But as not all cases are the same and statistics can't account for everything, I wonder if Elubas has a point. In a sense, 1300s very rarely meet super GMs, but I think at a certain point the super GM has to know the 1300 on a certain level. When I was at the level of a 1300, my super GM teacher had taught me so well that he had to steer clear of certain positions or I would have the upper hand! After all the years I still haven't beaten him, but we've also played comparatively few games against each other. However, I'll give you a story he told me that may shock you and also seems quite intriguing. My teacher's name is Matthew Looks, and he used to work at the Manhatten chess club. So one day, he's at the club and he sees this young russian kid sitting there studying. He's the best player in the world under 16, but he offers the kid a game. The kid is easily just a 1300 and here's my teacher, one of the best and youngest IMs in the world. First week the kid goes 0 and 10, but my teacher sees something in the kid's game. So every week the kid would come back and play my teacher. Months go by, every week they play. And if my teacher played the kid ten games he'd go 0 and 10, next week 1 and 10, next week 2 and 10, next week 3 and 10. Years go by and that kid(No longer a kid but an adult) has just won the world blitz championship. He even beat Kasparov. That kid? That was Maxim Dlugy, you might know the name.