What Does It Take To Win an Official Tournament Here?

Say for instance, Teary wanted to win a official chess.com, 1601-1800 rated tournament. How good would Teary have to be to actually have a chance of winning such a tournament?

If his rating was an established and stable 1799 (near the highest rating theoretically allowed), do you think he would have any chance at all?

Or, is it a given that a person with an established rating actually within the required rating range has virtually no chance of winning? (or conversely, that the only people with legitimate chances of winning big tournaments are new members whose real skill levels are far above the limit of the section they enter?)

Or I suppose, if this was made into a poll of sorts, then the question could be asked: how high of of a true rating (a measure of your actual skill on this site) would you theoretically need to have a chance in a 1600-1800 section?

1800?

1900?

2000?

2100?

2200?

higher?

By the time a player reaches the 1601-1800 range they should be reasonably reliably rated, much more so than the zero games completed premium members in the 1001-1200 range tournaments. Still some have not played enough to have a reliable rating and others have the audacity to improve their play over the course of 18 to 24 months the tournament takes.

In response to your question, higher than 2200.

Around 2400 strength would give a player a good chance to win a 1600-1800 official tournament.

746,928 Lats is the going rate

A Power Rating of 100,000

While in Tourneys .. Ish seem to do relatively fine against our "betters" or "equals" and then get beat to death by the lowers

I don't see why that should be the case at all. If you win your first game and it's against an 1880 or something, your rating will be in the high 1600s. That doesn't tell you much at all about the real strength - was it a good game that he got lucky in, or is he a master who had a relatively weak opponent for his first game? You can't know at all, it's just one game. I don't see why five or ten games would, by necessity, make it any more reliable, especially if they win them all.

The thing is though, a player with a real strength of 2000+ can start on this site fresh and reach the 1600's with just a handful of wins. And that is where a lot of the complaints lie.

Strong players are reaching mid level ratings within 5-10 games and are then signing up for all these tournaments. But since they don't really reach their 'natural' peak ratings until many games later, they keep on rising dramatically all through the tournament and end up way out of everybody elses' league by the end (heck, even Teary hasn't reached his true, natural rating yet and he has played over 30 full games!)

Teary has said it many times before: if chess.com would up the minimum games requirement to 20 or 25, then it would weed out of lot of the underrated players who are still trying to reach their natural peak.

ha, yeah, exepct that a lot of times, the majority of the "upper half" is rated way above the limit of the tournament itself. So even if Teary was rated 1800 in a 1600-1800 section, do you really think that would be good enough against hoards of 2000+ rated players in the same section?

Well, take a look at the 9th Chess.com Tournament, 1801-2000 section:  http://www.chess.com/tournaments/pairings.html?id=16399&round=3

In that tournament, the people advancing to the final round are rated 2253, 2225, and 2338.  Average of these is 2272, or 272 points above the max tournament rating.  I don't think that's unreasonable.  People do improve, and the fact that you're winning the tournament is going to improve your rating in and of itself.  A player truly rated 2000 would have a small chance to win against that, I think, but a chance.

Then compare the 1201-1400 section:  http://www.chess.com/tournaments/pairings.html?id=16396&round=4

In this one, the people playing in the final round are rated 2068, 2291, 1800, 1881, and 1965. Average rating is 2001, which is 601 points above the tournament maximum.  That doesn't seem reasonable.  A true 1400 player would not have a chance to win this.

Yeah, I would say something is broken here.  Maybe... I don't know, make sure that people have played at least 20 rated games before they can enter a tournament?  I hate to restrict a feature like that, but a system where people 600 points higher are in a section doesn't seem right.

When they got into them, they were 1601-1800, but then went up during the course of the tournament.

I hope not to come across as condescending, but thinking about this in terms of ratings is the wrong way to approach this question, in my opinion. What it takes to win an official tournament is winning. You need to try as hard as you can to win as many games as possible. If you do that, you have as good a chance as anyone.

But still, improving your rating by 400-600 points in the span of just a few months is in no way 'natural.' That kind of growth is indicative of an under-rated player still trying to reach his plateau, and not a player who is just studying and improving his skills. For example, can you imagine a stable 1400 USCF player increasing to 2000+ in just a few months? It would be silly.

But considering that it does take a bit longer for even an under-rated player to reach higher ratings, perhaps a sliding scale for tournaments would be more helpful?

1201-1400 Section - 20 game minimum (since the lower sections would have the most dramatic gaps between the limit and actual player ratings).

1400-1600 Section - 25 or 30 game minimum

1601-1800 Section - 30 to 40 game minimum

1800+ Sections - 50 game minimum.

I considered the sliding scale before your post, and had to reject it... Say you need 20 games to enter 1401-1600 and 25 to enter 1601-1800.  If you have 20 games played and a 1590 rating and sign up for the chess.com tournament, and then win 2 games, now you're suddenly ineligible for ALL of the sections, because you played MORE games.

Editing to add:  It's also the lower sections that are more affected by this, so a sliding scale probably isn't needed.  Like I showed before, the problem is over twice as large in the lower sections.

But that is the entire point of the sliding scale: to stop the super fast risers from getting in weak sections, and to allow the natural risers to get in the appropriate sections.

A person who is improving their game naturally will logically have their rating increasing slowly and steadily, which in turn means that it should take them more games to reach a certain section. So if anything, the 20-25 game requirement should be easy for the natural risers to meet.

But if a player is a super fast riser, i.e., an under-rated player trying to rapidly reach their peak rating, then the 20-25 game limit (or higher depending on the section) will catch them and prevent them from sandbagging.

The trick of course, would be determining what the average and natural increase in a player's rating is over time, so that appropriate game requirements can be set.

As for your 1590 example:

1. That issue wouldn't effect any player for very long anyways. They may miss one tournament (just as a player who currently only has 3 games would miss an official tournament now), but once they get past the 25 game mark (or the highest requirement there is for any section), then it won't ever be a problem for them again.

2. And honestly, even 50 games for correspondence chess isn't that tough to do. Even playing a few hundred games at once is perfectly doable for the average person.

I find that the quality of the games suffer when I play more than about a dozen at once.  

But anyway, we have to consider that, by imposing limits on who can join tournaments, we're taking a feature away from legitimate users of the site.  We should NOT limit in borderline cases, I think.  

The current policy is 5 games or be a premium member.  All this really does is ensure that someone doesn't join a tournament on their first day on the site and then never come back - it does very little to address the issue stated in this topic.  5 games is too few (and 0 in the case of premium members is WAY too few), but 50 games seems like too many.  10 to 20 would be more appropriate.  Much more than that, and you're overly restricting people from using the site.

We don't need absolute protection from underrated players.  Something to catch the more obvious ones would be fine.

https://www.chess.com/tournaments/pairings.html?id=16399&round=3

https://www.chess.com/tournaments/pairings.html?id=16396&round=4