Tournament Sections

Author: Mark Kaprielian

January 1998

I.                    How many Sections should we have for our Tournaments

A.            Experiences when we had two sections and averaged 23 players (1995)

1.             Splitting the field in two left both sections rather anemic.  In four round tournament, the ideal minimum would be 16 players.  When you have less than that, the pairings can become rather odd and difficult.  Also, due to the distribution of players appearing for the first round, what seems like the middle point at which to split the group into two sections the first week would often appear as a bad choice in the weeks that follow.  Essentially, 23 players with a random distribution were not sufficient to hold two good sections.

B.            Experiences when we had one section and averaged 26 players (1996)

1.             Due to the uneven distribution of players appearing for the first round, to avoid the problems of the previous year, we would more often than not, eliminate the second section and have just one, starting on the first night.

2.             Having one section would result in the new to chess or lower half of the field having a 300 to 500 point disadvantage for the first two rounds.  While undoubtedly educational for the lower ranked player, it was clearly, from casual comments heard afterwards, not encouraging experiences.  Many times I was asked if they couldn't somehow be paired with people closer to themselves.

3.             With the first two rounds being rather uneven in terms of ratings, the higher end of the field generally had to wait until the last two rounds to be in control of their fate, so to speak.  Often, because of the uneven distribution of ratings, several of the higher rated players would be paired significantly easier than others in point group.  This had the effect of making the order of finish seem like it was heavily influenced by luck of the draw instead of the contenders meeting head to head.

4.             In efforts to promote attendance in our events to players not yet participating at the club, I was told on several occasions of people who did not come to the club because of our tendency to combine sections.  Specifically that they did not want "easy" games for two or three rounds.

C.            Experiences when we had two sections and averaged 34 players (1997)

1.             The club averaged 34 players from an active pool of about 70.  The distribution of those 34 or so who showed to play would typically allow us to split the field into the two sections at a 1600 or 1650 rating with both sections being close in size.

2.             The two sections seemed to offer players in both sections, reasonably good competition often starting in the early rounds.  The disparity between ratings was not nearly as great as it had been in the past.

3.             The pairings in the later rounds were much better behaved than had been in the past.  The contenders for first place generally met by round 4 and their fates were in their own hands.

4.             These conditions seemed to be satisfactory to most participants.

D.            Experiences when we have two sections with 40 or more players (Currently)

1.             In both sections, but particularly the lower section, the disparity of the ratings is once again making for very uneven matches.  The lower section feels this difficulty more due to there being a steady stream of unrated players just beginning their competitive experience.

E.            Skewing factors

1.             The large number of people in the lower section and in particular the under 1400 level may be an aberration of the last few months.  Since September of 1997, an average of about two new to USCF players has joined the club.  A fair number of these players may well end up with ratings above 1400 after they complete their provisional rating phase. 

F.            Recommendation for the future assuming turnouts of 40 or more.

1.             Make all our regular time control events at least two sections for the following reasons:

a)             This gives people the choice of playing up or not.  When they want to test themselves, they can play up.

b)             It keeps the competition sharper than if just one section.

2.             Continue to have two sections for the next six months and see if the trends and distribution of players still merits splitting into three sections.  If so then:

a)             Organize our tournaments as three sections events.  If we had split the field for the January 1998 event on the first night of play, assuming no one played up, our sections would have been as follows:

Open      19 players

U1600    14 players

U1400    16 players

b)             Advertise our tournaments as three sections with the lower sections being combined if sufficient turn out or distribution does not occur.

3.             The rating distribution will have to be carefully observed to determine the likely population of each section.  We may wish to hold off on the three-section split until we can support sections of Open, U1800 and U1600.  It is my estimation that the number of people who would wish to play up when these sections are offered would generally be small as compared to other rating breaks.