The CCA can (and does) assign floors for their tournaments. As was pointed out earlier on this thread, the CCA is a business, and Bill can do business any way he likes. Some (though not all) other organizers also choose to use this list.
RE: Bill Goichber’s authority to assign ratings comes from rule "28E. Assigned Ratings for rated players. The director may assign a rating to any rated player."
This rule also appeared in the 4th edition of the rulebook. That means it has been around for a long time now.
Tim
I agree. The rule book is the authority! From my part, I was only wondering why 2000 and not something lower or closer to the U1600 that Todd won. We do make mistakes and that is also a possibility. I guess Todd will have to ask the CCA that, if he is interested in doing so. Be that as it may, any player may always elect to play or not to play if he or she disagrees with the assigned rating. This, I think, is the answer to the original post – it should be up to the organizers and TDs to try to prevent the sandbagging problem, not the USCF. Certainly, with the systems and information developed by people like Nolan and the rules and guidelines of the USCF Rule Book, we can make better decisions, but it is still up to us, organizers and directors, to police up tournament problems. That’s my opinion, anyway…
The problem I see with the CCA rating floor … (true) the rating floor is different than the USCF rating floor … the problem with the different rating floors, it only make an impact with the class players. Can understand how someone can find having a USCF rating floor … and a CCA rating floor as a way to stop a repeat of the same person to win the prize a second time.
If the CCA gave a 2400+ player a CCA rating floor of 2800, would it work(?), would the player(s) show up(?), it is not going to happen. If the CCA gave a 2600+ player(s) a CCA rating floor of 3000, would it work(?), would the player show up(?), it is not going to happen. Makes you think, they would not have a problem with the USCF 1600 players … would not even dream to do the same with a USCF 2400 players or the USCF 2600 players … so we are not so equal after all in the eyes of the CCA.
CCA is a private organization, so it can do whatever it wants, and often does. USCF has to be a little more algorithmic.
It is highly unlikely that somebody could win a $20,000 under-1600 prize without being at least a 2000-strength player. So I think that floor is reasonable.
In fact, that would be a good way for USCF to establish its floors. At the risk of suggesting complexity when everybody else wants simplicity, why shouldn’t the floor depend on the dollar amount? One class up for $1000-$4999, two classes up for $5000-$9999, etc (or something along these lines).
Bill Smythe
I disagree.
A 400-point rating difference corresponds to (roughly) 9 out of 10 odds in favor of the higher-rated player. So the odds of the higher winning 7 in a row would be 9/10 to the 7th power, which is slightly less than 50%.
If one player beats another 7-0, odds are the difference in playing strengths is much larger than 400.
Bill Smythe
To Bill Smythe: Yes, I do have a 1600 USCF floor now.
I can explain the 2000 CCA floor.
By the time I had returned home from Minnesota, I had already received a few emails from friends informing me of it. I took a quick look at the other HB Global winners and noticed that that their CCA floors were equivalent to their soon-to-be USCF floors. In other words, they’d each been bumped up one section, whereas I’d been bumped up three.
I was curious why I was singled out, so I asked Bill Goichberg. He responded that he had done some research and discovered my results in college chess tournaments (unrated ones) and my online results, and he felt those results justified 2000.
I was disappointed by that, but as others have pointed out, he is certainly allowed to assign anyone whatever ratings he likes. And in my opinion, he should be allowed to do so.
If the 1900 is having a good day/tournament it actually shouldn’t be that hard. Like you have already proven with the statistics, a 1900 can go 7-0 against 1500s about half the time. Which is pretty much a coinflip, but in my personal experience, I very rarely see 400+ point upsets in major tournaments (such as the US Open). These 400 point upsets usually seem to happen only in club events where the higher player might relax a little too much and get caught off guard.
So I’ll have to agree with franclg on this one. Also, over a period of 4 weeks, I had a 12 game training match (unrated G/60) with a teammate rated 310 points lower than me (I’m 1670, he’s 1360), and I won 11.5-0.5 (courtesy draw in the last game). So IMO, I think a 7-0 result against people 400 points lower is actually much higher than just a coinflip.
In regards to the “improving during a long layoff period” messages: After a 27 (or so) year layoff, I started playing again. Although Mike Nolan was able to find my old rating (above and beyond the call of duty in my opinion), after the ratings program changes over the last year or so I noticed that my rating was provisional. Now, I certainly don’t mind, and think it even makes sense. I assume it was done intentionally, but I’m not sure.
So… how do organizers deal with provisionally rated players with respect to prizes? A quick glance through the TLA doesn’t show anything. In an ideal world, I can see organizers wanting to reduce prize money for provisional ratings,but I would expect it to impact being able to draw in new folks negatively. Note that at the rate I get to play in tournaments, it’ll take me more than 2 years to get that P off my rating.
Just curious.
That is true. How many directors or I should say how many players have been to a tournament … when the top rated player lose their first game … and withdraws from the tournament.
Sure, have seen 400 point flips before … even a 800 point flip. Even had a master that should have lost to a class D player … only ending in a draw as he watched his own flag fall … both flags down – draw! But when you come to a tournament 35 minutes late for a G/60 … than spend 15 minutes eating and talking with a friend … only gives you 10 minutes at the board. Told the class D player … all you had to do is say ‘flag’ … not just watch it fall with a grin on our face.
I have to go with Bill on this. I think 7-0 would be roughly a coin flip for a player who is actually playing at their rating.
Back when I was going to US Opens, in my first five (in the 80s) I was paired down six times against players about 500 points below me (five first rounds, one second round). I was a bit off in one of those five tournaments and blundered and lost the first round game.
For that matter I should have lost the second round game as well, but my opponent blundered into a loss after I refused his draw offer when he had a clear two extra pawns in a somewhat complex ending. I would have lost that ending to most players my strength. Hubris kept me from accepting the draw offer and I was fortunate to not get burned for it.
I wasn’t sick, had slept enough, and didn’t arrive late. I just blew some moves in those two games and was fortunate to get a split.
Expert strength players will occasionally make blunders, and those blunders will some of the time be exploited by lower rated players. One thing that reduces that exploitation percentage is that many lower rated players don’t even look for an expert’s blunders and thus let chances slip through their fingers (the most dramatic case for me was a U.S. Open in a later round where an A-player missed that my opening blunder allowed a discovered pin/smothered mate in one when it wasn’t even move 10 - I ended up winning that in a complex middle game/ending).
When playing people at my club that are 400 points below me I expect a roughly even chance to go 7-0. Since some of those players are improving, and have not yet had their rating quite catch up to their strength, I would not expect much more than a roughly even chance. I would, however, expect a better percentage when playing the same person seven times, as I feel I would have the better chance of taking advantage of the other player’s repeatedly displayed tendencies.
Jeff and Bill,
I understand all that about going 7-0, but, to stay on the subject of this thread, my point is that a 1900 should be able to dominate a bunch of 1500s and win the section. Organize a tournament, have a big EF, big prizes, and then tell the guys in the U1600 that you are allowing a 1900 to play in the section! Explain to them that the 1900 really only has a 50% chance of winning! Yeah, right, they will buy that! They will scream bloody murder, and rightly so. A 1900 is a 1900 for a reason – the same reason why a 1500 is 400 points lower, and the former a lot better player. Besides, as we know, it normally does not take a 7-0 record to win a seven round tournament. Having said all that, I still like the chances of a 1900 going 7-0.
As far as Todd being assigned that 2000 rating floor by CCA after winning the U1600 section, I only suggested the possibility of an error by CCA. Again, we all make mistakes. I know I do! Todd mentioned that Bill researched his unrated college games and his online results. Not sure how, without his (Todd’s) help unless he (Bill) was also TDing or organizing those unrated events. Maybe he was… As far as on line results, I am guessing that he does not mean USCL Quick Rated events. ICC or something like that, I suppose. But, how can Bill get those without knowing the log-in name? Maybe Todd volunteered a lot of information and wanted a higher floor. I don’t know but, again, it still does not make sense to me. Be that as it may, yes, I agree, CCA has the authority to assign whatever rating they want and then it is up to the player to play (or not).
Interesting…I was glancing through the new Chess Life at our chess club last night and saw (but not read in detail) an article on this very subject – cheating. I think it was a letter submitted by IM Danny Kopek and others.
Todd is a Senior Admin at the Free Internet Chess Server (FICS) and as such a well-known player in that community. His handle there is not a big secret (and his real name is in his finger notes), so Bill G. would not have to dig very deep to find him. No big mystery there.
/Jens
In general, provisionally rated players are allowed to compete for all prizes as if they were established. I believe at the HB Global they were required to compete in either the Open or the Unrated sections. We had a provisionally rated player from our area play up there and be able to play up one class because he had enough games to give him a higher unofficial established rating. I believe that if provisionally rated players can’t win full prizes, that must be stated in pre-tournament publicity.
Alex Relyea
Big limits on prizes seems to be the rule for provisionally rated players in CCA events. In the upcoming world open, for example, a provisionally rated class E player will be limited to $2000. (unless playing up in a higher class)
No, I want the lowest floor possible. I volunteered no information. Bill had already found everything he needed and set the 2000 floor before I even made it back home from Minnesota.
My online handles are fairly well-known in the internet chess community (I’ve been an administrator on a major chess server since 1998), and you can even find them by Googling my real name.
My college chess tourney results can also be found by Googling.
I’m guessing that’s how Bill did it.
Edit: I guess I should’ve read the whole thread before responding to franclg. It’s already been spelled out by jmadsen.
Yep, page 8 of the new Chess Life issue. A letter from a bunch of players including GM Andrew Soltis, IM Danny Kopec, and Chess Life writer Jon Jacobs. More details at seniorchess.zoomshare.com
I have wondered for some time whether actual results with a 400 point rating difference match the expected performance (around 91%).
It would be an interesting though non-trivial task to check to see how well the expected performance formula matches up with actual results.
BTW, Australia is now using the Glicko-2 rating system, which may be the direction the USCF goes at some point in the future. I don’t know if it will help with sandbagging issues, though.
For more information on Glicko-2, see math.bu.edu/people/mg/ratings/glicko2desc.pdf
Of course. But we’re talking about a 1500-rated player who is really 1900 strength.
A 1900-strength player might be able to do well in a rinky-dink under-1600 section with small prizes. But he is likely to perform less well in a big-money tournament.
Why? Because big-money class tournaments attract under-rated players in droves. Some are sandbaggers, others are fast-improving, others – etc etc etc.
And if our 1900-strength player is doing well, he will be paired against these under-rated players, because they’re the ones that will have scores comparable to his.
In other words, he’ll have to do well against OTHER players who are also 1900 strength, if he hopes to win.
Bill Smythe
Those who present this line of reasoning clearly do not wish to follow it to its logical conclusion.
Who should be allowed to win, say, an U1600 prize? A lower-rated player who has improved rapidly? Nope, he’s “really” too strong. A player who used to be higher-rated but has fallen back? Nope, must be a sandbagger. The only ones who would be eligible would be “life C players.”
Now, this is a perfectly rational idea. (Whether it would be commercially viable is another question.) But those who advocate it should recognize that they are rewarding stagnation.