Forums

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

Sort:
Elubas
Shippen wrote:
Reb wrote:

Define " beat " .  If a 1300 wins the game because A) his 2700 opponent dies one move from delivering mate did the 1300 " beat " him ?  Or B) does the 1300 actually have to outplay the 2700 and not win the game through some freak occurence as described or horrific blunder from the 2700 ?  If A then yes its possible but if B then no its not possible .  I have won games in which I did not beat/outplay my opponent and I have lost games in which I didnt get outplayed/beaten .  

Basically the premise should be could a 2700 be outplayed by a 1300; then clearly not by the very definition of the rating gap. Unless that 1300 has just started playing and has not reached their optimum rating.

What is your definition of a rating gap? So a 2099 can never outplay a 2100?

Shippen
Elubas wrote:
Shippen wrote:
Reb wrote:

Define " beat " .  If a 1300 wins the game because A) his 2700 opponent dies one move from delivering mate did the 1300 " beat " him ?  Or B) does the 1300 actually have to outplay the 2700 and not win the game through some freak occurence as described or horrific blunder from the 2700 ?  If A then yes its possible but if B then no its not possible .  I have won games in which I did not beat/outplay my opponent and I have lost games in which I didnt get outplayed/beaten .  

Basically the premise should be could a 2700 be outplayed by a 1300; then clearly not by the very definition of the rating gap. Unless that 1300 has just started playing and has not reached their optimum rating.

What is your definition of a rating gap? So a 2099 can never outplay a 2100?

The rating gap of 1300 and 2700 is a vast gulf if those ratings are a correct assessment of their playing strength.

DavidPeters2

Wow, still going! It's funny because I don't think anyone disagrees the chances are microscopically tiny, the debate seems to be about whether that counts or not.

Players aren't robots, their standard of play varies. So a 2700 player might lose to a 2400, a1300 may beat a 1700, it can happen. If you had a gm on a really really bad day and a dream day for the patzer then it is possible. Never going to actually happen realistically (they never play) but there is a chance.

The OP says 'any chance' not 'is it likely', so the answer is yes.

Thread over and out please. 

 

Shippen

Needs locking

Elubas

Remember that ratings come from results. The reason why a 2700 is rated differently from a 1300 is because his results have been very different. That's where ratings come from, so that's all they mean. A rating difference isn't like some physical law that disallows certain results. In the end it's just going to be probabilities.

Elubas

I think people have said here that we are talking about a human 2700, not a computer 2700, and a human 1300, not a computer 1300.

But I'd argue that that point isn't even relevant. If we want to stay true to what a rating means, then the right way to program a computer 2700 would be to make it so that it will lose (or get two draws) to the 1300 approximately every 3000 games or whatever it was. If you did such programming, and had two computers battle each other out over millions of games, then one computer should get to 2700 and the other to 1300.

Elubas

And I found out through the OP's (RetiFan's) history that this thread has over 800,000 views! It almost can't fit all those numbers in the display. That is awesome. Let's get this to a million views.

Elubas

It often goes a few weeks or so without new posts, but it never truly dies. And I doubt it will. There is always someone who is going to bump a thread like this (who knows, maybe it'll be me some day), or someone who gets click baited and then makes a "he'll win if the other dude is drunk" joke.

Elubas

It's funny though cause like a few months ago there was a guy, I think his name was scrumpy jack or something, who bumped the thread saying how ridiculous it was and that it should be dead. Then why the hell didn't you let it stay dead lol.

tondeaf

This is really a good question. I am looking forward to many more posts until we get a definitive answer...

Ziryab

There is a definite answer. There is a wee chance. Your odds are better with the lottery. Vastly better. It's not even close. Winning the lottery is almost a sure bet in comparison.

Elubas

Winning the lottery would be way harder. Do you know how hard it is to win the lottery?

JayStanz

The 1300 player doesnt stand a chance. He will lose 100 out of 100 games, and maybe all in less than 25 moves. His best hope is that he will bore the GM into not paying attention. 1300 players dont know how to open well and give themselves a good chance for activity and attack, and they unwittingly hang pieces all the time. It was true of all of us at that rating. This is my humble and respectful opinion. 

DjonniDerevnja

One  1300 will lose 100 out of 100, but what about 10 000    1300`s will they lose all  1 000 000 out of a million? 

There is no doubt that shit might happen, and that a 1300 on a lucky day can play good enough to punish something. 1300 FIDE isnt that bad, maybe its ca 1650 online on chess.com? (I am 1461 FIDE/1773online)

Many normal people (maybe not GMs) do tilt after a blunder, and might play a bad move after the blunder too. I get punchdrunk after a blunder. Many GMs probably are so mentally fit that they can play well even when punchdrunk.

Elubas

"His best hope is that he will bore the GM into not paying attention."

And that's totally legit. That kind of psychology actually does occur in games with such large rating differences such as +-400.

mdinnerspace

2 types of people :

1. Those that believe anything is possible

2. Those that live in reality

Sred
mdinnerspace wrote:

2 types of people :

1. Those that believe anything is possible

2. Those that live in reality

And that's totally not the point. The point is that an event with a probability p>0 will eventually occur if the experiment is repeated sufficiently often, even if p is very very small.

joblinclitoo
In time odds, or piece odds your chances are 50/50. In a normal, rated game, 0.001% chance. Still not 0% though.
mdinnerspace

Wrong sred. My point is valid.

The point is in reality it would never happen.

Mathematics be dammed. Go ahead and believe the monkey/typewriter / Shakespeare idea if you want, Multiverses and all that hogwash. In practice it would never happen. 0.0000.... chanch. Come up with all the mathematical proofs you want. They only prove the formula is correct, not the reality of human experiance.

It's the same as asking could you ever beat the fastest ever in a 100 meter race? The answer is No.

Not ever.

It is absurd to argue that he may trip, was drunk or whatever. What's the point of using what ifs in a discussion of this nature?

Megabyte
joblinclitoo wrote:
In time odds, or piece odds your chances are 50/50. In a normal, rated game, 0.001% chance. Still not 0% though.

50/50?! Wow, that's extremely optimistic. What are the time odds? 10 seconds GM for 1 hour C-class player?