Promoting Chess via Tournament Design

Running Chess Tournaments
1

Mike I’m starting a new topic because my questions in the Web Design topic really weren’t on topic or even in the right section.

I was talking about scholastic or juniors purchasing the one day tournament membership. And your last post basically said that they can but it doesn’t make any sense to do so etc.

Warning-I’m going to go long winded here. :slight_smile:

First here is the end results of my idea.
gpcf.net/tournamentinfo/july … tional.htm

At our last Peoria Tornado (Which are game 80 4 round regular swisses)
I was unhappy that there was a 5 way tie for class c/d, a three way tie for class a/b, a two way tie for 1st over all. So I started thinking about changing our format. I suggested to fellow club officers we put it on our agenda to maybe change the January Tornado. That would have been the first opportune time to make changes. (That’s another problem if you want to try something out it takes 6 monthes to see how it works then another 6 monthes to adjust to that.)

Anyway at the time I was thinking that in a four round swiss one round is pretty much wasted, you are paired way up or down and the results are pretty much a forgone conclusion. So I started to think of ways to eliminate that game.

Then I was convinced that some players like that game, it is akin to having a paid chess lesson for the lower rated player.

So then I decided to leave our normal format alone.

But for a long time I have been saying that locally one of our major failings has been letting our Novice type tournaments wither away.

So we come to the sectional format idea which I’m sure has been done at some time in the past. Almost everything has.

But in my mind this format solves some problems.
1st The time devoted to a one day tournament is shortened. Enough so that you can extend the individual game’s time controls if you should wish too. In our first experiment with this format we are going to go with game 45, Start at 1 and be done by 6 at least.

2nd I think the major tieage(sp maybe not even a word) will be reduced due to the fact that there will be 8 or 6 player sections. I’ve added a top of the bottom half prize to give those who lose in round 1 a chance for something.

3rd The games should be more competive as far as playing players close to each other in rating.

4th It is a sneaky way of allowing Novice or Beginners a chance to play rated chess with out getting thrown to the wolves and spending a fortune to do so.
(This is actually why I was asking about the $10 adult membership. We are going to run this tournament at a loss. If you look at the prizes $60,$20 for and 8 man section and a $10 dollar entry fee you will see there isn’t anything left over for rating fees. But local TDs will do this for free, we have a free site and we are only advertising via word of mouth. The other expense is we are offering to pay the $10 dollar one day fee for newcomers.

I don’t really expect to be overwhelmed by newcomers. But I wanted to know if I had the option to use the $10 amount on a junior etc if we decided we wanted to. )

5th a tournament like this can be organized on the spur of the moment. Things can be tried and changed very quickly. We can get real feedback.

I know there can be problems with dropouts and we have a plan, but the hope is that since this is a short event most players will play all the games.

The other problem is that everyone must be there at the beginning of play otherwise they won’t be paired in.

We will have one houseman available at the start of the tournament in order to even out the numbers. A possible second houseman to help in handling any dropouts.

If we get dropouts in two different sections we will give each player that is due a bye a full point and offer to let them play each other for ratings only. If we get 3 or an odd number of dropouts that is where the other houseman comes in.

Ok having said all that, if anyone wants to come and play in Peoria and bring a friend that has never played rated chess before you are more than welcome.

2

If you’re concerned about the pairings in the early rounds of your events, there are several ways to address that:

  1. Use multiple sections restricted by ratings. Someone who chooses to play up several classes knows what he’s getting into.

  2. Use accelerated pairings.

If you’re concerned about the pairings of the last few rounds not giving you clear class prize winners, you can take that into account when pairing those rounds, though that might impact the overall winners.

I honestly believe that if kids (or their parents) aren’t willing to pay $25 or less for a full year membership, then they really aren’t ready to play rated chess anyway. Have unrated sections to get them started. Fairly soon they’ll want to know how they compare against other players, and that’s when they will begin to have some interest in USCF membership and a USCF rating.

3

I have long felt that accelerated pairings are highly over-rated.

The first round feels like the second round. The second round feels like the third round for players in the top or bottom quarters, and like another second round for those in the middle.

The third round works well only for players in the top or bottom eighths. For the vast three-fourths in the middle, it feels like the first round.

Better just to have multiple sections.

Another idea, to avoid foregone-conslusion pairings, is 1-vs-2 pairings. In round 1, pair 1 vs 2, 3 vs 4, 5 vs 6, etc. In subsequent rounds, pair by score, but within each score group pair 1 vs 2, 3 vs 4, etc.

Bill Smythe

4

So here are the results from our experimental Sectional Tournament.

gpcf.net/allresults/2006/Wee … chart.html

We had just as many players as our last Tornado Tournament. We had a good local turnout. In fact the top section consisted of all Peoria players. (Actually it looks like all but 5 were from Peoria, The other 5 being from the Bloomington area.)

We didn’t have any dropout problems to deal with. Some of the weaknesses of this format discovered was the inability to seperate players from the same family and maintain the ratings order. (I did change the sections from the predicted 886 to 868 to keep two brothers in different sections. But it just happened to work that way. The top section had 3 Leali’s)

1st place was won by someone in the top half of each section but it would have been possible for someone in the bottom half to take first and ice out the top half prize wise. (But maybe if the top half is going to let that happen they deserve it.) Still had massive ties in the top of the bottom half. (But that doesn’t bother me as much here.)

On the positive side we started at 1pm and were done just a touch after 6pm. 1st place was won outright in each section. We had two players play that weren’t USCF. One of them opted for the 6 month USCF membership. We gave back 110 percent in prizes and memberships.

5

<<<2. Use accelerated pairings>>>

My experince with single accelarated parings was that they just largerly switch the 1st and 2rd rounds in terms of non-competitive games. There is some advantage to this (a few more upsets & etc) but not much. I don’t have any meaningful experience with double accelerated which may be a better solution. But I guess that it mighty switch the 1st and 3rd rounds. I could be wrong on this.

John Barnard ran a few experimental tournaments where in each round 1 played 3, 2 played 4, 5 played 7 and etc. Necessary adjustments where made for colors.

Top players would always hate this system because they have to play their main rivals right off the bat.

For our standard local tournament, 5 round Swiss, one or two experts, lots of 1900-1500 and some beginners, the players just loved the system. Most mid-range players want to play up improve thier ratings rather than down and the beginners get to play a couple of real games among themselves before moning up.

6

Can you describe “double accelerated pairings”?

Bill Smythe

7

First two rounds accelerated.

8

If you take a 32-player four round tournament and all of the games are decisive with no upsets then with standard pairings (ignoring colors) player 12 would play in sequence 28, 4, 20, 6 and player 24 would play 8, 32, 16, 18. Acceleration would have 12 playing 4, 20, 24 and 5 while 24 would play 32, 16, 12, 17. Acceleration shifted the round 2 opponents to round 1 and the round 3 opponents to round 2. Player 12 also plays 28 and 6 with standard pairings versus 24 and 5 accelerated, and player 24 also plays 8 and 18 standard versus 12 and 17 accelerated.

1 would play 17, 9, 5 and 3 standard and 9, 5, 3 and 2 accelerated while 32 would play 16, 24, 28 and 30 standard and 24, 28, 30, 31 accelerated.

For the middle, round three accelerated looks vaguely like round one standard and both round fours look quite similar. Thus, for players in the middle, acceleration has only a moderate effect on the pairings while the extremes are get significantly different pairings with acceleration.

A major argument for acceleration is that it cuts down on the number of players in the extreme score groups. There is a less-used argument that it reduces the number of major mis-matches. The major argument is viable (barring people doing ridiculous things like using acceleration in a class tournament), but the less-used argument seems to only significantly apply to the extremes.

9

Um, that’s not double-accelerated, that’s just regular accelerated.

Round 1 is paired by quarters. In round 2, top half winners play each other, as do bottom half losers. Top half losers play bottom half winners.

Did you have something else in mind for either “double” or “single” accelerated? If so, please let us know.

Bill Smythe

10

Anything to add? Nah!

The “problem” we were trying to solve was fixed real easy by breaking the tournament into two sections and allowing the ambitious to play up.