Today I ran a scholastic tournament along lines suggested to me by Rob Jones: a rated JTP section for players in third grade and below, unrated sections for older players. Here’s what happened:
Only three players showed up for the K–3 section. Two of them were brothers. My initial solution was to change the format from four-round Swiss to double round robin, something I’d done once before.
The older brother, a rated player who didn’t want to play against his sibling, asked if he could play up to the 4–6 section. I decided to allow this.
This, of course, left only two players in the K–3 section. For them, a “double round robin” consisted of two games, with colors reversed for the second game. The same player won both games. Both players are USCF members but do not yet have any rated games.
Then I tried to submit this sad little thing for rating. Submitting it in its initial form returned the error, “This appears to be a match – change section type.” But changing it to a match returned the error, “Both players must have established ratings for a match to be ratable.” Keeping it as a JTP section and adding the third player back in as a “withdrawn player” with two unplayed games accomplished nothing.
Have I created a situation where the games played simply cannot be submitted for rating by any means? I doubt that the kids care much; they both went home with trophies. But their parents might inquire.
I believe that you can have a “match-like” section that is not a match with office approval. I’d imagine that you’ll have to wait until Monday to get this approved, but a tournament in which only two players show up is definitely not a match.
There’s no obvious way to differentiate between an event that had bad turnout and a match between two players, so we assume any section that has 2 or more games between just two players is a match and must meet the match restrictions regarding player ratings. The office can override those restrictions, but the event will still be rated as a match, eg, subject to the match limits on ratings gains and losses.
What would happen if you included the name of the third player (the one who played up) in the lower section, but with no games played? (He would still be listed in the higher section also, of course.) Maybe this would get around the rating software.
I think the tournament validation program already eliminates players with no played games, and I have proposed a more comprehensive set of rules defining which events must be treated as a match so that TDs can’t try to bypass the match rules with events that contain sections in which players play multiple games, but for the most part just against one opponent.
See BINFO 201100044 for the latest draft of those rules, which the Executive Director and Executive Board have yet to take under consideration.
A “match” between two players who are USCF members but no previously rated games would be utterly meaningless until one or both of them play more games. I am not sure it is a good thing for these games to be shoe-horned into the rating database. It seems rather reasonable that there are special rules about the ratings of players in matches. Perhaps you shouldn’t really be trying to come up with workarounds for getting this section into the ratings database. Even if the rating formulas spat out some rating for the two players, what could those ratings possibly mean?
Maybe it would be best to wait until one or both of these players have more rated games against players with real ratings. Then perhaps you can find a way to get these games into the system. Meanwhile, maybe it would be best to forget about rating these games. Would that mean refunding money to the two players?
We rate tournaments in which every player in it is unrated. And with rerating, unless you fudge the dates of the event, the system will try to rate it in chronological sequence anyway, so there’s no point to waiting.
The restrictions on matches go back many years, though they probably weren’t as mechanistically enforced in the past.
To me this is a reasonable exception, because you can’t predict your turnout, and most organizers hate to tell players that they can’t play rated games just because turnout is low.
Mostly those limitations are there to try to prevent ratings manipulation, though I suspect that there are some mathematical underpinnings as well. As I recall from Elo’s book, having games against a large pool of opponents is one of the underlying assumptions. A few games against the same opponent shouldn’t make a big difference.
It would not be meaningless. It would be relevant to future ratings calculations that one of these two players had beaten the other one twice.
Or, if you think about it another way, every rating is utterly meaningless until a player plays more games, because the rating is not used for anything except section placement and pairing.
Is the event in your TD/A workarea? You can submit a membership exemption request (category ‘uscf office action’), explain the situation and request a waiver. The office normally tries to process those within a day.
Today’s the cutoff for the March list, if you get your exception request in quickly we should still be able to get it rated today.
I left the message on the voice mail of Walt Brown, who called me back about 15 minutes ago. The office overrode the errors, and the event has been submitted successfully.