Forums

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

Sort:
chawit1234

Last fide rated tournament ,a 1300 beaten 2 ,2200+ rated

SmyslovFan

Let's give two mutually exclusive possible scenarios and say which is more likely:

A) a legit 1300 defeats a 2700 in a standard classical time control FIDE rated game that matters to both players at least once, ever.

B) That never happens.

Both are possible, but are mutually exclusive. Which is more likely?

For me, it's B.

SmyslovFan

And btw, for the umpteenth time, by legit we aren't talking about someone who is underrated or cheating.

imsighked2

There is a chance that an unseen comet stikes the Earth tomorrow, but I won't bet on it.

SmyslovFan

Besides, we aren't talking about infinities. FIDE won't be around that long, neither will chess, neither will humanity.

mdinnerspace

Why not less than 0? People start throwing out numbers wily-nily. As, this number exists, therefor that number is possible, proving the third number is conceivable. Less than 0 is an irrational number, want to talk mathematics?

matethedead

A 1300 player can easily beat a 2700 player... with a baseball bat... before the game... then, he'll beat him at chess as well.

matethedead

... with the chess board.

mdinnerspace

Why not less than 0? People start throwing out numbers willy-nilly. As, this number exists, therefor that number is possible, proving the third number is conceivable. Less than 0 is an irrational number, want to talk mathematics?

jechi7

"Less than 0" could be negative one or negative 456/1000. Those are RATIONAL numbers. And yes, I do want to talk mathematics!

 

 

Good bye!YellYellYellYellYellYellYellYellYell

fayfay1

Sure this has been said, but FIDE uses an Elo system with a standard devation of 2000/7.  Thus the mathematical probablity of a 1300 rated player beating 2700 rated player is 

xz[ |x1 - x2| / σ ] = xz[ | 2700-1300 | / (2000/7) ] = 4.792845 x 10^-7

or about 1 in 2086443.  Meaning that you would exepect the lower rated player to win on average 1 game out of 2 million.  Of course this model does not work perfectly and doesn't take in to account the individual's rating confidence interval.  But mathmatically speaking there is a chance.

mdinnerspace

People want to believe anything is possible. This conclusion is not based on mathematical equations, but is simply a philosophy so to speak. Mathematical infinities are then employed to support a philosophical belief.

fayfay1
mdinnerspace wrote:

People want to believe anything is possible. This conclusion is not based on mathematical equations, but is simply a philosophy so to speak. Mathematical infinities are then employed to support a philosophical belief.

The question is entirely mathmatical, because the rating system is math.  The prompt is not asking if I could beat Magnus, but if a 1300 player could beat a 2700.  By using the rating system in the question you are pulling it away from an opinionated question into a question of math.  I am merely giving the answer based on how the rating system works.  

Of course the rating system is inaccurate.  Everyone in chess knows not to take ratings to seriously.  Ratings say little to nothing about the style of play or the player themselves, it merely sums up trends in the relative and historic preformances against other players.

 

So could I beat Magnus Carlsen-not in a million years -opinion.

What is the probality (chance) that a 1300 beats a 2700 - ~1 in 2 million

Robert_New_Alekhine

A 0.00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 chance.

Ziryab
imsighked2 wrote:

There is a chance that an unseen comet stikes the Earth tomorrow, but I won't bet on it.

I'll bet against it every day until hell freezes over, knowing that I will always collect and even on that rare day when I lose, I will not need to pay.

Ziryab
fayfay1 wrote:

Sure this has been said, but FIDE uses an Elo system with a standard devation of 2000/7.  Thus the mathematical probablity of a 1300 rated player beating 2700 rated player is 

xz[ |x1 - x2| / σ ] = xz[ | 2700-1300 | / (2000/7) ] = 4.792845 x 10^-7

or about 1 in 2086443.  Meaning that you would exepect the lower rated player to win on average 1 game out of 2 million.  Of course this model does not work perfectly and doesn't take in to account the individual's rating confidence interval.  But mathmatically speaking there is a chance.

You may need to go back 100 pages or so, but data was presented in this thread showing that when the Elo difference exceeds 500, the higher rated player tends to overperform statistically.

mdinnerspace

So what are the chances the sun won't rise tomorrow? Get out your calculator. What are the odds? There is always a chanch right? People just don't get it. Statistical chances are NOT proof of anything being possible.

fayfay1

@Ziryab Ah I see I figured it was posted somewhere in this tread but I haven't read all 166 pages yet.  Smile

mdinnerspace

The universe is x years old. It is expected to last for y years. Chances are the sun won't rise tomorrow is z. However small the odds are, there remains the possibilty. The math never lies. BS. But it does play games with your head.

mdinnerspace

Numbers can not take into account certainty. Laws govern the universe. They can not be broken. No matter how you jiggle the numbers.