Forums

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

Sort:
Elubas
JeffGreen333 wrote:
Elubas wrote

Don't tell me you're getting my position mixed up with the gambler's fallacy! If I say that there is a decent shot of a 1300 winning once out of 10,000, I'm not saying that his chances are really low the first 9,999 games and then his odds are really high on the 10,000th. I'm saying that the chance of a 1300 to win on at least one of those games, which could be the first game, the 800th game or the 6500th game, is not so bad. 

 

After playing 9,000 games against a GM, that 1300 player might become a 1700 or 1800 player, with all of that experience.  Then, after he has improved his playing strength and has learned all of the GM's opening repertoire, I think he could win one or two games out of the last 1,000.  So actually, his chances do become better on the 10,000'th attempt, not due to the odds getting better, but due to the fact that he will become better than 1300 strength after playing that many games against a GM.  In the case of the lottery, each ticket is equal no matter how many you buy, but here, the player would get better over time.  Quite the paradox.  

Of course I know that, which is why I'm controlling for that, done many times in this thread. So for example instead of imagining 10,000 games in a row between two players, we could imagine 10,000 2700s playing 10,000 1300s. I'd count on an upset somewhere with at least one of those games. That's a reasonably conservative estimate, given that the rating system suggests something like a 3 in 10000 chance for the 1300 to get a point. And yeah you can say that the formula breaks down for large rating differences, but in practice it breaks down in favor of the weaker player. With huge differences the stronger player often overestimates his chances and tends to score worse than the formula would suggest.

KairavJoshi

Ask Vishy Anand directly if he thinks a 1300 can beat a 2700 during his live lesson in the Prodigy Program this November.

https://www.chess.com/blog/GeniusKJ/massive-discount-for-november-prodigy-program-with-vishy-anand

JeffGreen333
Elubas wrote:
 

Of course I know that, which is why I'm controlling for that, done many times in this thread. So for example instead of imagining 10,000 games in a row between two players, we could imagine 10,000 2700s playing 10,000 1300s. I'd count on an upset somewhere with at least one of those games. That's a reasonably conservative estimate, given that the rating system suggests something like a 3 in 10000 chance for the 1300 to get a point. And yeah you can say that the formula breaks down for large rating differences, but in practice it breaks down in favor of the weaker player. With huge differences the stronger player often overestimates his chances and tends to score worse than the formula would suggest.

Well, if there were ten thousand 2700-rated players in the world, most of them would probably be Russians.  We all know how they like their vodka, so maybe they would see no harm in getting drunk before playing a lowly 1300 player.  So, they might make a few drunken, careless moves and blunder a piece.  lol   So, maybe the odds are even better than 10,000:3.   I'm trying to think outside of the box, here.  tongue.png

JeffGreen333
GeniusKJ wrote:

Ask Vishy Anand directly if he thinks a 1300 can beat a 2700 during his live lesson in the Prodigy Program this November.

https://www.chess.com/blog/GeniusKJ/massive-discount-for-november-prodigy-program-with-vishy-anand

Can a 1300 beat a 2700?  Yes, of course.  Anything is possible.  Will he?  No.   lol   

BronsteinPawn

Yeah, the OP is stupid, everything is possible. WHAT A STUPID QUESTION. If this question didnt exist chess.com wouldnt had have to buy so much new servers to storage all this data.

Elubas
JeffGreen333 wrote:
GeniusKJ wrote:

Ask Vishy Anand directly if he thinks a 1300 can beat a 2700 during his live lesson in the Prodigy Program this November.

https://www.chess.com/blog/GeniusKJ/massive-discount-for-november-prodigy-program-with-vishy-anand

Can a 1300 beat a 2700?  Yes, of course.  Anything is possible.  Will he?  No.   lol   

Though it can be the case that a 1300 will beat a 2700.

DjonniDerevnja

I did beat a 1300 yesterday. He started fantastic with a beautiful najdorf and came pawn up. Then he played a very strong and fine middlegame, with one active rook working hard- two pawns up. In the endgame I managed to activice my rooks  and leveled it out.In move 55 it probably was a draw, but he thought he should queen first, and ran for it, letting my pawn run too. He overestimeated his own speed, and made one error- going with the pawn before his king- I slided my king to his queeningsqare, and he resigned.

Next time he will improve his endgaming, and grow into a more complete player. A 1500+. He was clearly better than a 1465 in the first 40 moves.

The 1300s are decent , talented chessplayers, and most of them are kids.

JeffGreen333
DjonniDerevnja wrote:

I did beat a 1300 yesterday. He started fantastic with a beautiful najdorf and came pawn up. Then he played a very strong and fine middlegame, with one active rook working hard- two pawns up. In the endgame I managed to activice my rooks  and leveled it out.In move 55 it probably was a draw, but he thought he should queen first, and ran for it, letting my pawn run too. He overestimeated his own speed, and made one error- going with the pawn before his king- I slided my king to his queeningsqare, and he resigned.

Next time he will improve his endgaming, and grow into a more complete player. A 1500+. He was clearly better than a 1465 in the first 40 moves.

The 1300s are decent , talented chessplayers, and most of them are kids.

Yes, they often attack me early on and I have to defend.   However, after I defend correctly and their attack fizzles out, they are down significant material and it's an easy endgame win for me.  

DjonniDerevnja
JeffGreen333 wrote:
DjonniDerevnja wrote:

I did beat a 1300 yesterday. He started fantastic with a beautiful najdorf and came pawn up. Then he played a very strong and fine middlegame, with one active rook working hard- two pawns up. In the endgame I managed to activice my rooks  and leveled it out.In move 55 it probably was a draw, but he thought he should queen first, and ran for it, letting my pawn run too. He overestimeated his own speed, and made one error- going with the pawn before his king- I slided my king to his queeningsqare, and he resigned.

Next time he will improve his endgaming, and grow into a more complete player. A 1500+. He was clearly better than a 1465 in the first 40 moves.

The 1300s are decent , talented chessplayers, and most of them are kids.

Yes, they often attack me early on and I have to defend.   However, after I defend correctly and their attack fizzles out, they are down significant material and it's an easy endgame win for me.  

;y 1300 opponents doeant attack to early. The last one was basically smarter and better than me, especially in opening play. He3 had a won game after 40 moves, but it was only his late middlegame and endgzame that was worse. His problem probably is that he is so good at chess that he usually decides his games in 45 moves, and gets little endgame-pracis.

Magnus as a kid was the other way around. He was so bad at openings that he got forced to learn how to survive bad positions, and  struggle him self  into the game, then finish them off in the endgame. Just like I did on Sunday, but of course Magnus was fighting mastermonsters, when I only met a 1300 child.

My point, in this whole debate, is that 1300 fide is a lot harder to beat than most non-otb tournamentplayers thinks. 1300 fide is ca the same strenght as 1700 online-chess.com. Of course 1300 is far too far away from 2700, but they are strong.

JeffGreen333
DjonniDerevnja wrote:
My point, in this whole debate, is that 1300 fide is a lot harder to beat than most non-otb tournamentplayers thinks. 1300 fide is ca the same strenght as 1700 online-chess.com. Of course 1300 is far too far away from 2700, but they are strong.

I wouldn't go that far.  Maybe a 1300 FIDE is about 1450-1500 on chess.com (Glicko).   Kids are usually much better than their ratings are though, since they improve faster than their ratings can keep up with.  So, he might be a 1700 Glicko player by now, this is true.  

DjonniDerevnja
JeffGreen333 wrote:
DjonniDerevnja wrote:
My point, in this whole debate, is that 1300 fide is a lot harder to beat than most non-otb tournamentplayers thinks. 1300 fide is ca the same strenght as 1700 online-chess.com. Of course 1300 is far too far away from 2700, but they are strong.

I wouldn't go that far.  Maybe a 1300 FIDE is about 1450-1500 on chess.com (Glicko).   Kids are usually much better than their ratings are though, since they improve faster than their ratings can keep up with.  So, he might be a 1700 Glicko player by now, this is true.  

I am  a middleaged 1465 Fide, and 1720 online. I feel my chances are much better against 1600 online than 1300 fide longchess. 

You are right about those kids playing better than their rating. I was very happy when I defeated that 14 yo 1345 on sunday. Except for his endgame he felt like 1600+.

JeffGreen333
DjonniDerevnja wrote:
JeffGreen333 wrote:

I wouldn't go that far.  Maybe a 1300 FIDE is about 1450-1500 on chess.com (Glicko).   Kids are usually much better than their ratings are though, since they improve faster than their ratings can keep up with.  So, he might be a 1700 Glicko player by now, this is true.  

I am  a middleaged 1465 Fide, and 1720 online. I feel my chances are much better against 1600 online than 1300 fide longchess. 

You are right about those kids playing better than their rating. I was very happy when I defeated that 14 yo 1345 on sunday. Except for his endgame he felt like 1600+.

Well, the endgame is a big part of chess, especially when you move up in class.  So, if his endgame is weak, that might explain why he's only rated 1300 FIDE.   Ratings are accelerated on chess.com though, because of two factors:  1. Most people play a lot more games on here than they play OTB tournament games, so their rating is more up to date.  2. The Glicko (chess.com) rating system allows for larger and quicker changes in your rating, compared to FIDE, USCF or ELO.   I am a 1470 USCF player, but my actual playing strength now is probably around 1700-1800.  None of my ratings have caught up to my current strength yet, since I study chess more than I play these days.  I don't pay too much attention to a player's rating when I decide on my opening or my style of play though.   A Grandmaster could be playing his very first game on chess.com and still be rated at 1200.   So, I certainly don't take any 1200's lightly.   

Elubas

"I am a 1470 USCF player, but my actual playing strength now is probably around 1700-1800."

And I'm an expert but my playing strength is probably as if I were an international master.

DjonniDerevnja
JeffGreen333 wrote:
DjonniDerevnja wrote:
JeffGreen333 wrote:

I wouldn't go that far.  Maybe a 1300 FIDE is about 1450-1500 on chess.com (Glicko).   Kids are usually much better than their ratings are though, since they improve faster than their ratings can keep up with.  So, he might be a 1700 Glicko player by now, this is true.  

I am  a middleaged 1465 Fide, and 1720 online. I feel my chances are much better against 1600 online than 1300 fide longchess. 

You are right about those kids playing better than their rating. I was very happy when I defeated that 14 yo 1345 on sunday. Except for his endgame he felt like 1600+.

Well, the endgame is a big part of chess, especially when you move up in class.  So, if his endgame is weak, that might explain why he's only rated 1300 FIDE

His endgame was not close to the very good openingplay and fine middlegame. I think that goes for many kids. If they play a lot blitz, the openings gets a lot work, and the games are often finished or decided on time before a long endgame.  A 1300 is often a good player with large holes in the reportoire. My opponent will grow a lot strenght working on the endgame.

TheAuthority

Bump

aj415

www.chess.com/forum/view/game-analysis/how-to-thread-how-to-scalp-players-rated-230022002100-2000-as-a-1300

how about 2300??!

DjonniDerevnja
Lasker1900 wrote:

The fact that 2700-rated grandmasters routinely run up 100% scores in simultaneous exhibitions against 30 or more 1500-2000 rated players might give you some hint

I dont think all the 2700s is capable of 100% simultanscores every day. When GM Jon Ludvig Hammer was at 2700 or 2695 he played simultan in my club. His score was 11,5-1,5 against the kids (rating from none to 2100), and 18-4 against the adults (rating ca 1200-2100). The winning kid was Timur Gusarev, unrated, maybe nine years old.

Even Magnus can lose in simultan. He lost to  Jens Erik Rudolph (1891) in a 70 game simultan.

http://www.nordstrandsjakk.no/?p=3366

aj415
DjonniDerevnja wrote:
Lasker1900 wrote:

The fact that 2700-rated grandmasters routinely run up 100% scores in simultaneous exhibitions against 30 or more 1500-2000 rated players might give you some hint

I dont think all the 2700s is capable of 100% similtanscores. When GM Jon Ludvig Hammer was at 2700 or 2695 he played simultan in my club. His score was 11,5-1,5 against the kids (rating from none to 2100), and 18-4 against the adults (rating ca 1200-2100). The winning kid was Timur Gusarev, unrated, maybe nine years old.

http://www.nordstrandsjakk.no/?p=3366

Well being unrated we don't know what his strength actually was so thats a wildcard but good link very insightful. pretty cool

coicidiently before someone bumped this thread I just made a thread on this topic since i am only rated 1300 but i compiled my best wins on here over 2000+ including a 2300 an IM and a CM in blitz . check it out and leave a comment!

www.chess.com/forum/view/game-analysis/how-to-thread-how-to-scalp-players-rated-230022002100-2000-as-a-1300

DjonniDerevnja
aj415 wrote:
DjonniDerevnja wrote:
Lasker1900 wrote:

The fact that 2700-rated grandmasters routinely run up 100% scores in simultaneous exhibitions against 30 or more 1500-2000 rated players might give you some hint

I dont think all the 2700s is capable of 100% similtanscores. When GM Jon Ludvig Hammer was at 2700 or 2695 he played simultan in my club. His score was 11,5-1,5 against the kids (rating from none to 2100), and 18-4 against the adults (rating ca 1200-2100). The winning kid was Timur Gusarev, unrated, maybe nine years old.

http://www.nordstrandsjakk.no/?p=3366

Well being unrated we don't know what his strength actually was so thats a wildcard but good link very insightful. pretty cool

coicidiently before someone bumped this thread I just made a thread on this topic since i am only rated 1300 but i compiled my best wins on here over 2000+ including a 2300 an IM and a CM in blitz . check it out and leave a comment!

www.chess.com/forum/view/game-analysis/how-to-thread-how-to-scalp-players-rated-230022002100-2000-as-a-1300

It was very unexcpected that Timur was the one to take him down. Timur is a smart little kid, but there were stronger kids that lost.

It was impressing that Hammer was able to defeat Andreas Garberg Tryggestad in a simultan. The next year they met in a lonchessgame at Arendal Grand Prix, and it was a draw. In that tournament Andreas, Jon Ludvig and GM Kjetil Lie was top three , undefeated with 4 points each.

I guess Hammer put so much energy against the superkids, that to little focus was left for the smallest.

A nine year old unrated kid is absolutetly strong enough to punish a blunder in the late middlegame and set a mate in three, no matter how strong the GM is. 

DjonniDerevnja
Lasker1900 wrote:

I had the pleasure of watching GM Nakamura sweep a field of 38 players, most of them rated quite a bit over 1300, in about two hours. In the opening, he took about 5 seconds at each board, yet at least 10 of his opponents were in serious trouble before move 10. The difference in class was overwhelming.

Many GMs play second best moves in simuls, to give the kids a chance. This is very sporting, and I applaud it, but it does distort the results

Nakamura, Carlsen and Gristchuk are very difficult to beat in simultans, that for them has blitztimecontrols and long or rapid for the opponents. They are the best blitzers, and plays blitz att master longchess-strenght.