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

So that might suggest that the chances are even higher than 3 out of 10000, not lower :)

Well the question isn't, would you favor the 1300 to win, it's, could the 1300 win despite the odds against it.

:-(

If 2700 rated player is very drunk there is certainly a chance 

And here I thought they all began a game with a shot or two.

I have heard that 50 milligram is the correct measure if you shal perform on top, but do anybody know if it is 50 milligram alcohol (96%) or 50 milligram vodka (40%) ? I know about a GM that drank more than that before facing Magnus. He opened well, got and advantage, but got unpresice in the endgame and lost. This proofs that when drinking alcohol a 2550 can lose to a 2850, but says nothing about a 1300 versus a 2700. 

40 % alcohol is my motto before a game. Always drink and play chess responsibly.

50 % alcohol is reserved for post match analysis.

96 % alcohol is impossible !!

Well, if you were to drink it, wouldn't it be fatal ?

40 % = 80 proof

50% = 100 proof

The problem is if a 1300 rated player plays the amount of games required according to the equation to get a win or even a draw, his playing strength should be significantly more than 1300 by that time.

A probability is not an "amount of games required."

Pure grain alcohol is 100% or 200 proof, there's 151 rum and Everclear is (i think) 180 proof.

I've been as high as 2100 strength and there is ZERO chance I could beat a 2700 in a rated game where he didn't have to leave for an emergency. A 2400 would be a small percentage chance of a draw. Possibly .01% or .015% of winning. Maybe less :0

With a difference of 300 rating points the higher rated player is expected to score ~80% So in a 5 game match a draw or two is not unlikely.

Carlsen drew with a 2400 a year or two ago. He drew with a few 2500s in the Olympiad earlier this month.

It is a mistake, a huge one, to make the assumption that given more chanches, the odds turn in your favor for the lower rated to win. Statistical properties can be applied to a roulette wheel, but not to winning a chess game vs a much higher rated opponent.

This has been my arguement. A mathamatical formula DOES NOT represent true chanches. You can play all the games you want, the more games does Not increase the chanches of the lower rated winning. He will lose everytime, no matter the amount of games. And please.... not this bs about more games, maybe a heart attack etc. Don't be absurd. Be real :)

The odds don't change of course as long as playing strength is the same. That's a well-known fact of probability that I do not deny. My point is by playing many games most people will improve some so their chances then would be better than they were before. So in a way the chances of this happening are asymptotically close to zero so that you can safely say it will not happen under normal conditions.

Your proposition is it's impossible. Not that it's very very unlikely, but that it's completely impossible.

But you give no argument. Only that you feel like it shouldn't happen the way you feel like elephants shouldn't fly (it's about your feelings because chess and elephants are unrelated).

You are correct, there is no mathematical formula... because all you need to do to find the odds are to count the possible outcomes. You said earlier in this topic (#4959) that it's a mistake to assume the possible outcomes are possible... and I think that pretty much speaks for itself

I really don't see much point in arguing over "impossible". I take this position just to be a hard head, I suppose, to the many posts that think a realistic chanch can be expecteded. I shout impossible! And it sure gets the blood flowing! Anyway, the topic was quite dead before I began :)

You have to admit, the thread moved on to theoretical territory (not the chess type!) with statistics, infinities, multiverses, absurdities, close encounters, comparisons etc., etc. Not to mention all the formulas! Some educational, some way out in the ozone !