How many games would a complete beginner have to play against a grandmaster to win?

Sort:
Avatar of davidkimchi

I read somewhere if someone is a 400 rating higher than you, they have a 90 percent or more chance of winning.

So a grandmaster with about 2600 Elo would easily beat a 2200 Elo player 90 times out of 100

How about if its a grandmaster vs a complete beginner who has played for a few days.

How many games do you think it takes for the beginner to win, given both try their absolute best

Avatar of justbefair

A million billion.

Avatar of PekarchikArtemiy

:) :) :)

Avatar of neatgreatfire

until the GM dies of old age

Avatar of Kraig

According to the Elo Win Probability Calculator, the expected score of a 2500 rated player vs a 1000 rated player in a classical game is 0.999999924.

I believe this means the beginner wins 1 in every 10,000,000 games.

In blitz and bullet, anything can happen. I scored a fluke win when I was a 1500 vs a 2600 GM in 3 minute blitz (I won on time in a slightly worse position).
So if its blitz or bullet, realistically maybe 1 in 1,000 to 10,000 games.

Avatar of Cobra2721

In classical the GM wins every game

Avatar of Inkblot23

Considering a rating of 200-350, ignoring time, and with the fact that the GM would genuinely want to crush you, your chances are at a rough estimate of less than a hundredth percent.

Avatar of HARDILAZIZDHILLON

PLAY TILL THE GM DOES NOT SAY ,'' ENOUGH!! I RESIGN, TAKE AWAY ALL MY POINTS".

Avatar of Cobra2721
Inkblot23 wrote:

Considering a rating of 200-350, ignoring time, and with the fact that the GM would genuinely want to crush you, your chances are at a rough estimate of less than a hundredth percent.

 

Avatar of MARattigan
davidkimchi wrote:

I read somewhere if someone is a 400 rating higher than you, they have a 90 percent or more chance of winning.

So a grandmaster with about 2600 Elo would easily beat a 2200 Elo player 90 times out of 100

 

How about if its a grandmaster vs a complete beginner who has played for a few days.

How many games do you think it takes for the beginner to win, given both try their absolute best

Enough games to take years at any rate.

Exactly how many would depend on how the beginner improves in the meantime. He would at least be getting excellent practice.

Avatar of MARattigan

Would depend on whether the beginner had the flair of, say, Capablanca.

Avatar of davidkimchi
Theqmjmmasta wrote:

in classical if the GM actually wants to win, its probably not a question even if they play hundreds of games

Agreed, possibly even tens of thousands.

But there has to be a point where the GM makes a mistake, and the beginner gets better and even lucky at the particular game to win at least once or draw

Avatar of davidkimchi
NervesofButter wrote:
davidkimchi wrote:

I read somewhere if someone is a 400 rating higher than you, they have a 90 percent or more chance of winning.

So a grandmaster with about 2600 Elo would easily beat a 2200 Elo player 90 times out of 100

 

How about if its a grandmaster vs a complete beginner who has played for a few days.

How many games do you think it takes for the beginner to win, given both try their absolute best

You could play a billion games against a GM.  And if all youre doing is playing?  You wont beat the GM.

I think after 1 million games, a beginner can win at least once if they get lucky or the gm makes a blunder

Avatar of Cobra2721
davidkimchi wrote:
NervesofButter wrote:
davidkimchi wrote:

I read somewhere if someone is a 400 rating higher than you, they have a 90 percent or more chance of winning.

So a grandmaster with about 2600 Elo would easily beat a 2200 Elo player 90 times out of 100

 

How about if its a grandmaster vs a complete beginner who has played for a few days.

How many games do you think it takes for the beginner to win, given both try their absolute best

You could play a billion games against a GM.  And if all youre doing is playing?  You wont beat the GM.

I think after 1 million games, a beginner can win at least once if they get lucky or the gm makes a blunder

Nope

Avatar of Cobra2721

A GM crushes beginner in every game. Beginner NEVER wins. To be honest, it would take a 1500 like me a really long time, and a beginner can NEVER defeat a GM. End of argument.

Avatar of IpswichMatt

It's been said that if you give a chimp a typewriter and an infinite amount of time he will produce the works of Shakespeare.

https://en.wikipedia.org/wiki/Infinite_monkey_theorem

This seems like a similar problem to that, but much more likely. I think  Kraig above in post #6 had it about right.

Of course, this is ignoring the possibility that the beginner might become rather handy at chess after a few hundred thousand games against a GM

Avatar of MARattigan
NervesofButter wrote:
davidkimchi wrote:

I read somewhere if someone is a 400 rating higher than you, they have a 90 percent or more chance of winning.

So a grandmaster with about 2600 Elo would easily beat a 2200 Elo player 90 times out of 100

 

How about if its a grandmaster vs a complete beginner who has played for a few days.

How many games do you think it takes for the beginner to win, given both try their absolute best

You could play a billion games against a GM.  And if all youre doing is playing?  You wont beat the GM.

If that were invariably the case then GMs wouldn't beat GMs either, because none of them has played billions of games since they were beginners.

Avatar of davidkimchi
IpswichMatt wrote:

It's been said that if you give a chimp a typewriter and an infinite amount of time he will produce the works of Shakespeare.

https://en.wikipedia.org/wiki/Infinite_monkey_theorem

This seems like a similar problem to that, but much more likely. I think  Kraig above in post #6 had it about right.

Of course, this is ignoring the possibility that the beginner might become rather handy at chess after a few hundred thousand games against a GM

very interesting read, thanks for sharing

Avatar of Cobra2721
Theqmjmmasta wrote:

It would take an extremely long time for the beginner to prevail. Like extremely extremely long. And also the infinite monkey theorem works kind of on probability, its a bit different with chess

 

Avatar of IpswichMatt
Theqmjmmasta wrote:

It would take an extremely long time for the beginner to prevail. Like extremely extremely long. And also the infinite monkey theorem works kind of on probability, its a bit different with chess

Yes, a beginner would not just make random moves. Would that make the beginner more or less likely to win than a machine that just generates random moves? (I would have thought more likely, but I could be wrong)