As I prepare for my first organizing/directing experience, questions keep popping into my head. Sometimes, I post them to get help.
Today’s query: Is it problematic to have all unrated players (or a majority) playing in the same rated section? My first tournament will be quads, and I don’t know whether a few new people will be joining. My instinct is to put them into the bottom section, but I’m not so sure about that when I get several newcomers at once. I know I can assign a rating of say 1200 (which would be too high anyway), but if there are several unrateds, I see the same problem arising.
Having a section consisting mostly or entirely of unrated players (and, basically, each quad is its own section) is no problem. If you know a particular unrated player is strong enough that he should go in a higher quad, you as TD have the discretion to assign the player a rating based on that knowledge and to use that assigned rating to determine the quad in which the player competes. However, you should not assign an arbitrary rating of 1200 (or any other arbitrary rating) to players about whose playing strength you know nothing.
You might want to spread the unrateds across different sections, so that each section has at least one (preferably two or three) rated players. Of course, this could be difficult if you have no idea how strong the various unrateds are, or which ones are stronger.
If all players in a quad are unrated (and have no previous tournament history), the USCF computer, in calculating post-event ratings, will have to guess how to seed each player’s rating. And that guess could turn out to be way, way off.
Not really. The computer does that anyway, at 50 points per year, up to age 26, and using fractions, based on zero. I’ve noticed some CCA tournaments have fairly large sections that have all unrateds. The point is that the computer can handle it. Anyway, ratings based on 3 games are going to be awkward (to say the least).
OK, so the computer doesn’t guess. It has a specific algorithm for guessing, based on the players’ ages.
But a new player’s rating has a better chance of coming out closer to fact than fiction (IMHO) if he has at least one previously-rated opponent.
Granted, 3 games isn’t enough to be reliable anyway. In fact, ratings are not published until they are based on at least 4 games, although they’ll show up on the crosstable.
The spring national scholastics and the national open both have rated sections restricted to unrated players.
Some of those unrated players have actually played in a tournament and have an unofficial rating (often too recently generated to show up in even the following month’s supplement), thus giving a starting point for players in the tournament even rather than having every player actually start unrated.
P.S. As far as I know, accelerated pairings have never been used in them.
When I have had new unrated players come to play in a quad tournament, I thought it was best to give them three rated players to play against for their first event. When I did not know much about their strength I put the younger player in a lower rated quad and the older player in one of the middle quads. This gave them a better chance to score at least something in order to get a rating. If several unrated players were playing, I never put more than two in a particular quad.
In regular Swiss System events of four rounds, I made sure to get the unrateds as many rated players to play as possible. No unrated was given a full point bye, even if he deserved one, so that he could get a publishable rating as soon as possible.
Regardless of the type of tournament, I always made sure to thank each new player for coming to the tournament and put a “Chess Life” and a tournament flyer in his/her hand and invited them to come back. I was advised by a mentor long ago to always shake the hand of every player as they were leaving the hall at the end of the tournament and give them flyers. The personal attention pays dividends.
Bill Goichberg once told me about a high school event he ran with no (or almost no) rated players. He wanted to use accelerated pairings, but had no ratings for the “top losers vs bottom winners” round 2 pairings. So he paired the 12th grade losers against the 9th grade winners. Big mistake. The pairings backfired, and he ended up with 3/8 of the tournament at 2-0 (as opposed to 1/4 with standard pairings, or 1/8 with accelerated pairings that work).
It’s not a problem per se. What the rating program does (or did the last I heard) is assign unrated players who are 26 or older a rating of 1300 and younger unrated players a rating equal to 50 times their age in years and fractions of years. So it could rate a quad consisting entirely of unrated players, though the reliability of the ratings would obviously be poor.
The rulebook does not address the issue of what to do with unrated players for quads, though it does mention in rule 28A (for swisses) that “Some directors prefer to assign an arbitrary rating of 1200 or 1300. for pairing purposes only, to all unrated players.” I gather that the 1300 number was chosen because it is the rating that is initially given to unrated players by the rating program, and that 1200 is sometimes used because the directors expect that players who are new to tournament play may not be playing their best at their first tournament. Rule 28A continues, though, by noting that “Such ratings usually place them at or near the bottom,” and this would most probably not be true if this procedure were used at a kids’ tournament, where an assigned rating of 1200 or 1300 might place a player in the highest, or one of the highest, quads!
