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

You are absolutely right, and I am guessing. This tread is so longliving that I hope to reach 2000 before it dies out. Most players peaks at some level. My plan is to get above 2000, therfore it will happen. In my club I can see huge improvemnts among the kids, slow and steady improvemnts among the lowrated adults and most status quo among the 1600-2100 adults. I am joking about becoming FM when I reaches the age of 93. It is a goal, but I really doesnt see realism , and illness an death can slow me down.  It looks like that people that sets themselves goals tend to reach them.

Playing solely in a club, the occasional out of town tournament gives ratings in a fish bowl. You're comparing yourself to the same known players. The better players rating goes way up while the opposite happens for those who lose. A rude awaking often awaits players with a "high" rating who venture out of their small circle, while lower rated players may fare better than expected.

hi. i once read that every 200 points difference decreases your chance of winning against the higher rated player by 25 % . just my 2 cents.

Eh, usually the highest rated player at a local club is at a disadvantage and is somewhat underrated. The kind of ratings that make it so they can score 3.5 out of 4 and still lose a few rating points. The lower rated players get a bit of a boost.

When the highest rated player plays in a big event where they're in the middle of the pack, they can usually gain some rating points.

For a quick reference it's (roughly) like this:
100 points difference is about 65% chance for higher rated player
200 points is about 75%
300 is 85%
400 is 90%
higher and higher it just goes to 99.999...
Although for big rating differences it's not a reliable formula anymore.

With a small pool of players, there will always be a few players at the top. Their rating can be unreliable, as they continually beat the same players. When they match up vs established players from a much larger pool who have played a greater diversity of players and hence other styles and openings, it's was my experiance the small club player found it quite more difficult to win vs similiar ratings.

This was years ago, being from So.Cal and did very well as a class B, winning several big tournaments. I moved to Utah, with a very small pool of players. I easily won the State B Championship, found the B players were equivalent to C players from Cal.

When I played in Oklahoma, I'd meet players from Texas and California (no idea why California people would visit) and they'd often say we were all underrated 100-200 points heh.

I guess it just depends on the area.

With a small group of regulars who are all improving... their ratings wont improve if they only play each other. That's my guess for what happened.

Point is, the rating system is designed to establish a certain percentage of players that represent each class. If there were 10 C players all playing head to head, the winners rating will undoubtedly move higher to B, when in fact he should remain at the top of the C bracket.

I believe you are correct. Often it is difficult for a small group of players to establish a "real" rating. Lower rated may well in fact fare much better in different areas, but in general I think the hot shot player who wins regualary at his small club finds it more difficult in the bigger out of town tournaments. Always exceptions of course, maybe his competition was very weak, no way to tell his true strength.

The rating system doesn't spread players out unless they're scoring a certain % better though. If the 10 players were very close in skill, they can stay the same ratings.

I mean, with good days and bad days they could shift a little... and maybe after 1000s of games they'd be spread out, but this would take a long time, and more likely that someone is learning from their games while another person isn't.

I think the issue lies in the fact that players in a small group get to know each others game and style. The better players learn to adapt and figure out ways to beat the same player they've seen over and over. It can be a different story when they match up with players they have not seen before. Often it is more a matter of style and temperament in the lower classes.

Hmm, again my experience is a little bit the opposite

When I play a higher rated player over and over and over, I get to know their openings and tendencies really well, and I think it tends to even us out.

When a lower rated player plays me over and over and over I think they do better against me than otherwise.

I think because they go home and study our game, while I just need to do the same thing next time. Until eventually they're getting further and further into the game in an equal position with me.

---

But maybe my experience is just full of lazy higher rated players hehe. If they were preparing something a little different each time maybe the lower rated player would never get comfortable.

0110001101101000 wrote:

The rating system doesn't spread players out unless they're scoring a certain % better though. If the 10 players were very close in skill, they can stay the same ratings.

I mean, with good days and bad days they could shift a little... and maybe after 1000s of games they'd be spread out, but this would take a long time, and more likely that someone is learning from their games while another person isn't.

I don't see our opinions as being very far apart. One thing to keep in mind is a class is +/- 100 points. 1401 and 1599 is still class C. There can be quite a real difference in the skill set. What can happen with the system is the 1599 earns say 1650 while the 1401 falls below 1400 if there exists a small pool of players.

Depends on how much they're losing though. If the 1400 loses 7 out of 10 against 1599 his rating... goes up about 2 to 4 rating points

I think a player who has played 100 games vs 100 different opponents and is rated 1800 has an advantage, would be more likely to win(again, on average) than the player who has played 100 games vs 10 different opponents and earned the same 1800 rating. Interesting question, it's my opinion. Don't know of any study to verify results

I remember reading that the rating system allows for 1% of all players to call themselves a master. A 2200 rating or higher. If a pool of 100 players only played each other over say 10 years and one player consistantly won a great majority of the games, I assume he would achieve the master rating. Is he really a master? Depends on the real strength of the other 99. Maybe he should be higher/lower, no way of really knowing till he ventures out and plays other established masters.

It works both ways. If you had 100 master level players who only played each other maybe they're all rated 1300 and the best guy is 1500.

I often play a similar range of people based on the tournaments I go to and I don't experience any rude awakening when I play in ones with very different players.

It's like theoretically possible, the sort of thing you're describing, but it would take an absolutely extreme situation for it to happen. Even with a pool of 10 players you still should have plenty of variety -- you might be worse than player x, but better than player y, who is better than player z, etc. You're fine.

Come on. Respect science just a little. Don't just assert claims like that without even the tiniest ounce of scepticism or testing.

And that's just it. You have absolutely no idea of whether your theory is actually true. When you're just randomly putting together conspiracy theories, you might as well just flip a coin to decide whether you're right or not.

 We have a fake Fischer and a fake Kasparov. I think a fake Carlsen would be appropriate.