Yes. Very slim chance but yes. Usually the 2700 is expected to thrash the 1300 but I suppose if the 1300 has more energy going into the game, has prepared well, and plays the game blunder-free and the 2700 is tired and isn't playing well in the game, yes.
Is there any chance that a 1300 rated player can beat a 2700 rated player?
Sort:
FYI: Dharmik Banka claims to be a GM. FIDE shows him to have a performance rating of 964 after 5 games.
His current +2300 rating here is extremely impressive.
No.
MorraMeister wrote:
Based upon several responses, there are alot of players that clearly don't understand the rating system and how it works.
You can look it up on your own time, but Arpad ELO invented a mathematical model for ratings many years ago. Based upon the statistical distrubtion of ratings, the following chart will give you the exact odds of the lower rated player defeating the higher rated player.
Rating Difference / Probability
0 50.000% 100 35.994% 200 24.025% 300 15.098% 400 9.091% 500 5.324% 600 3.065% 700 1.747% 800 0.990% 900 0.559% 1000 0.315% 1100 0.178% 1200 0.100% 1300 0.056% 1400 0.032% 1500 0.018%I'm sorry, but that doesn't seem right. How can a player have a 50% chance to beat a player of equal rating ?
If that applies to both of them and each has a 50% chance to beat the other, then where are the draws ? 50% + 50% = 100% of games would be decisive then.
Also, regarding draws, the chance for a draw is probably higher in the higher rating ranges because in a game between two patzers there will be a higher material imbalance on average (due to blunders).
We are are susceptible to heart attacks; god forbid during a chess game. I for one ; would not the point.
Ok, got it now, thank you for explaining this in detail.
GMdharmikbanka23 wrote:
once I beat a 2721 rating player(my sir)when we were playing silly matches.My rating was 1287
You also NOT a GM and have cheated on chess.com and have been reported.... or maybe al those sub ten move games were just lucky?!
Elubas wrote:
"Indeed, it may be more likely a particular 2700 would lose than a particular 2300."
That seems extremely implausible. Unless you just mean that they have to have a heart attack to lose, in which case the 2700 and 2300 are in the same boat? Chess skill keeps you out of trouble, it makes you wary of the little pitfalls/blunders you can fall into, it triggers your chess alarm bells, etc. Having that kind of stuff at a much higher level (2700 vs 2300), will indeed be extremely helpful in preventing accidents.
The idea is more like this, a GM can't win a queen vs lone king position any better than I can.
Chess skill keeps you out of trouble... when the opponent applies pressure, is the end of that sentence. As for random accidents, there is no reason to believe that a 2700 strength player would have better accident prevention (in terms of losing to a 1300 rated player) than a 2300 rated player.
And no, of course people are not equally likely to have heart attacks!
Anyway, as I said, the math suggests this is true. That after a certain rating gap, skill isn't correlated with results. And IMO common sense also suggests this is true.
"there is no reason to believe that a 2700 strength player would have better accident prevention (in terms of losing to a 1300 rated player) than a 2300 rated player."
There is a ton of reason! Surely you would know a lot of what goes into chess strength is consistency. You learn how to not miss mates in 1 because they cost you the whole game. So when you say, well, there could be an accident, the problem is, accident avoidance is a rather large part of what playing strength represents in the first place.
So for me, even if I had a lapse in attention -- and missed a mate in 1 that would occur after a move I was considering -- probably other parts in the way I approach chess would keep me from actually making that move. I would eventually notice it before actually playing it. Well, stronger players than me have that same sort of thing but to an even greater extent. Maybe some day I will miss mate in 1 and that'll be crazy. It'll be even crazier, and happen less often, for a master to do so. And so on.
Basically, the kind of lapse that would have to happen in a 2700s brain for him to make a large enough accident would have to be much larger than what would have to happen for a 2300. A 2700 is so good that even a lapse in his brain still won't be enough for him to be close to missing a mate in 1, because other parts of his chess sense will kick in and catch the mistake.
"The idea is more like this, a GM can't win a queen vs lone king position any better than I can."
Well I would actually disagree, there. I think a GM is much less likely to randomly forget his queen was hanging and somehow mess up the mate. His instincts would keep him from doing it to a greater extent than you. You might mess up that mate once in five thousand times, whereas he might mess it up once in five hundred thousand times.
Can a GM be affected by the "higher rated player psychology?" Of course, but so is every player who is playing a weaker player. And the GM would be better equipped to cope with it because he has more ways in his thought process of avoiding blunders. The accident could still happen, but the better the player, the more extraordinary factors it would take to happen.
If you say "is there any chance" the answer is yes, no matter how small, a chance is a chance. If you say reasonable winning chances, then I'd say no.
"Anyway, as I said, the math suggests this is true. That after a certain rating gap, skill isn't correlated with results."
Is there a study to which you're referring? That could be legitimate.
"And IMO common sense also suggests this is true."
But if what I'm saying is wrong, it's definitely not because I wasn't following common sense. When you know what goes into chess skill, accident prevention becomes a part of you, and this will not stop happening when you hit 2300. I could grant you that it's not easy to tell the difference between the accident prevention abilities of a 2300 vs a 2700, especially if you're much weaker than 2300, but that doesn't mean the differences aren't there.
Thinking about it, I think I have to agree with you that a 2700 player will be more consistent. I'll have to change my argument to the difference in consistency is so small that it does not measurably affect the results due to other factors having a comparatively overwhelming affect such as mood, motivation, physical health (even just a headache), etc.
This is why I say a particular (not all, just one) 2700 player may fare worse than a particular 2300 player.
Elubas wrote:
"Anyway, as I said, the math suggests this is true. That after a certain rating gap, skill isn't correlated with results."
Is there a study to which you're referring? That could be legitimate.
Ok... so mathematically, you would have to say they're always positively correlated. Increase rating difference, then increase expected score.
I was saying we could just make a guess based on the shape of the curve since the Elo formula isn't intended for very large differences. The shape being asymptotically approaching 100%.
My idea (guess) was that if you actually plotted data points and tested for correlation, that you wouldn't notice when the rating difference is very large because other factors would matter so much more.
Those quoting the .018 are correct. That is how many points a person rated 1500 pts lower in the Elo system is likely to score out of 100. But they make a mistake: In chess, it's possible to score points with draws. The chance of a person with an established rating 1500 pts lower beating the higher rated player is much lower.
Elo's rating table isn't equiped to work out the number of wins a person may get, just the overall probable score.
Suffice it to say that .018% is too optimistic.
So, when rounding, the correct answer to the question is
0 Chance out of 1,000, and probably 0 chance out of 10.000.
When you calculate the number of serious games that have ever been played between a 1300 and a 2700, the answer becomes:
Astronomically small.
So I beat a GM at 3 2 one day in around 20 moves. He is a 2400 player and I am Class A.
Was he on drugs?
Yes, there is! Nothing is impossible.
If a player suffers a heart attack during a match and taken to the hospital, I assume he forfeits the match. The opponent has a +1 score even if about to be checkmated. Did he win the game? Rediculas question imo if a game would ever take place with a 1400 difference in rating in serious play. Pairing protocols make this impossible. Club games? Got a big enough bribe and the GM just lost the house at the poker tables, well maybe.
"Indeed, it may be more likely a particular 2700 would lose than a particular 2300."
That seems extremely implausible. Unless you just mean that they have to have a heart attack to lose, in which case the 2700 and 2300 are in the same boat? Chess skill keeps you out of trouble, it makes you wary of the little pitfalls/blunders you can fall into, it triggers your chess alarm bells, etc. Having that kind of stuff at a much higher level (2700 vs 2300), will indeed be extremely helpful in preventing accidents.