Bring Back the "Toss" in the Last Round!

Running Chess Tournaments
1

BRING BACK THE TOSS FOR COLORS IN THE LAST ROUND!

In the early 1960’s, when I began to consistently play in USCF rated
chess tournaments we had a simple method of determining color
allocation in the last round of any event when the opponents were due
to have the same color. If we read the word “(toss)” on the pairing
chart the two players would do exactly that; one of us would toss a
coin and the other would “call it”, heads or tails, to see who got
White in the last game. This was a manifestly fair way of deciding
the color situation and I do not recall a single instance where any
type of disagreement occurred after the “toss”. This also meant that
in a small tournament you could be fairly certain by studying and
correctly appraising the positions of your closest rival(s) during
the penultimate round exactly who you would have to play in the last
round, but NOT what color you would have. This is important because
nowadays the current rule that “the higher rated player (always) gets
his due color in the last round” grants some players a more perfect
knowledge of their last round pairing than they deserve, and can also
unfairly effect their decision making processes during the
next-to-last round game. Consider the following:

On December 19th, 2010 I played in a four round, game/60 USCF rated
tournament at Dean Ippolito’s very attractive club, the Dean of Chess
Academy in Branchberg, NJ. There were only ten players, which
included five rated over 2000, namely IM Dean Ippolito (2547), Karl
Dehmelt (2319), Fred Wilson (2095), Matthew O’Brian (2047) and Steve
Ferraro (2005). Dean and I each had Black in the first round and
won, while in the second round I drew with Dehmelt and Dean beat
O’Brian. In the third round Ippolito and I were black against
Dehmelt and Ferraro respectively. He drew fairly quickly using his
favorite Petroff’s Defense while I won a long, pawn-up ending against
Ferraro. I am not saying that Ippolito decided not to play harder
for a win because he could easily infer he would have white against
me in the last round, but with another, less ethical player this
could easily be the case. Now suppose I had no edge against Ferraro
but a time advantage, knowing for certain I have black against
Ippolito might encourage a less ethical player to try to beat Ferraro
on time, in order perhaps to get his entry fee and some “meal money”
back, and knowing that he MUST have black against the IM in the last
round. And before someone says that it wouldn’t matter what color a
“patzer” like me has against an IM I will point out that I drew Dean
with white two months earlier, and had very conscientiously prepared
against all three of Ippolito’s favorite defenses to 1. e4. Also,
the rating differences in this scenario could be much closer, with
the “Wilson” instead being a 2318 low rated IM!

What I am getting at is why should any player have such “perfect
knowledge” of his last round pairing? While I understand that you
can often figure out who your last round opponent must be, I do not
understand why both players, if they are due for the same color,
should know for an absolute certainty what color they will have. I
can only surmise that this intrusion into commonsense fair play was
created solely for the convenience of tournament directors. When we
used the “toss” the director probably had to use 5-10 minutes
learning the colors for some of the last round pairings, depending on
the size of the tournament. HOWEVER, I DO NOT THINK WE SHOULD
CONTINUE TO SACRIFICE FAIRNESS FOR THE SAKE OF EXPEDIENCY. And, if
you agree with me that the current USCF rule “higher rated player
gets his due color in the last round” is unfair then please urge the
delegates in your state chess association to vote to rescind it at
the next Delegate’s Convention in August, 2011, at the U.S. Open
Championship.

So, for the sake of fairness to all, BRING BACK THE “TOSS” IN THE LAST ROUND!

    Sincerely, Fred Wilson (USCF member, almost continuously,

since 1961; Chess Journalist of the Year 2003)

PS. Does anyone know if one included in the TLA for a tournament
that in the event of both players being due the same color a “toss”
would be used, rather than the current rule, if USCF could refuse to
rate this event? I would like to hear some senior TD’s opinion on
this.

2

I remember when the “toss” was considered to be the preferred way to deal with colors in the last round. Will have to double check old Harkness method, but I seem to recall that he felt this to be an acceptable decision process. Don’t remember anyone complaining about it either. Several TD’s who were at one time in the USCF hierarchy used to use the “toss” as simple and a fair way to deal with color problems. Some fussbudgets who like to have 1000+ page rulebooks must have pushed for the “due color” to higher rated idea.
Since the rating reports do not request color allocation, see no reason why these would not be rated. They always used to be. Only FIDE has special rules requiring the reporting of color allocation as well as result.

3

Higher rated (well, actually higher ranked where a 3.5-0.5 1900 is higher ranked than a 3-1 2250) was the rule before the current rulebook when both players were equally due for a color. The current edition also uses full-color history and only goes back to higher ranked if the round-by-round color history is identical (such as in the cited case).

4

The current rulebook already allows for a coin toss (choosing colors by lot). See rule variation 29E4c. on page 146 in the rulebook. This is no longer the most popular method of choosing color in the last round; thus, it is not seen much any longer.

5

I would think that this would be well within the realm of “major variation”, which means that it would be allowed if it were announced in all advance publicity. I’m not so sure if the organizer would get into trouble if he waited until the opening announcements (or even the last round) to announce this policy.

Alex Relyea

6

Pairing rules, including color assignments, are part of the Tournaments section of the rule book. The rule Alex is talking about applies to “Swiss System” tournaments. It is like truth in advertising. You can’t say a food is “organic” unless it complies with USDA-defined specific rules. But you can call a food “natural” or “whole” and those terms can mean whatever you say they mean, within reason.

So it is with the term, “Swiss System”. The USCF won’t let you call a tournament a “Swiss System” unless you follow its rules for SS tournaments, using only allowed variations, and announcing “major” variations in advance.

By the way, the definition of “major variation” is a variation which could affect a player’s decision to participate in the tournament. Since nobody can really divine what might affect a player’s decision to participate in a tournament, it is largely a matter of opinion as to whether a variation is “major”. There is no canonical list anywhere of variations that are deemed “major”, although if you scan through this forum, you will find lots of precedents for what other organizers have called “major” and lots of opinions about what might be “major variations”. Your guess is as good as anybody’s as to whether tossing for colors in the last round is a “major variation”.

Going further, the USCF rulebook doesn’t say that every USCF-rated tournament has to be a “Swiss System” or “Round Robin” tournament, which are the two main tournament structures covered by the rules. There can be, and are, other types of USCF-rated tournaments. For example, rated “Ladder” tournaments are rather common in clubs, and the USCF rulebook has nothing to say about how those are paired.

So call your tournament a “Wilson Toss System”, or any other name you fancy, and you can do just about anything you want as regards pairings, color assignments, etc. So long as the actual games are chess games, played according to the rules in the first, “Play of the Game” section of the rulebook; and all the games are between paid-up USCF members; and the TD is certified by the USCF, the USCF will rate the games of your tournament.

Of course, the players will want to know what the Wilson Toss System is. You will have to tell them somewhere, and that place should be easy for them to find. They will then decide for themselves whether they want to play in a Wilson Toss tournament.

By the way, in the same spirit of truth-in-advertising, I am not a Senior TD, so decide for yourself whether you want to pay any attention to this.

7

  1. As Tim Just pointed out, variation 29E4c provides for assigning color in the last round to players with equal priority for the same due color by lot. Personally, I don’t think this would be a “major variation” insofar as I don’t imagine players reconsidering a decision to enter a particular event because this variation is used. I think it would be sufficient to announce this variation at the opening announcements (and probably post a sign somewhere).

  2. As Mike Nolan has written before (though I risk paraphrasing): the rating system doesn’t care how the player’s opponent was decided. And the online rating report for USCF currently does not include color information, as Jeff Wiewel points out. There is no reason the rating report would be rejected just because colors in the last round were determined by lot.

8

. . .
Authoring or supporting a rule that needlessly gives one player an advantage over another seems like a snub of fundamental sportsmanship.

The “toss” is an improvement over biased rules that needlessly favor one player. However…

Better than having the players toss might be for — WinTD or SwissSys to randomly assign colors in these cases.

One strength of the random assignment approach is that the first player to arrive can set up the board without waiting for his tardy opponent. (The first player already knows his color because it is notated on the final round pairing sheet, as usual.)

Another advantage is that the TD’s system reliably has colors recorded for the players. In the foreseeable future, capture of color data in reports to the USCF could conceivable begin. There could be other unforeseen uses of the color data.
. . .

9

Fred,

It’s good to see you push wood again. You might not remember me, but I’ve bought books from and chatted with you at the USATE. I was the fat guy.

A few points of order: First, we mean “higher-ranked,” not “higher-rated,” determines who gets due color. Rank is determined first by score, then by rating among players with the same score.

Also: Karl Dehmelt is one of the nicest 2300 players I’ve met—it’s also good to see him push wood again—but he is an FM, not an IM. Dean is one of the nicest very strong Masters I have met; his dad is one of the real good guys in chess. I met them when Dean was 9 and I could still beat him now and then. Steve Ferrero, (note correct spelling), is absolutely honest; I have known and played against him since the '90s.

I don’t always say nice things about chessplayers I know. The guys mentioned here are all first-class in terms of what you are talking about, even if as a hypothetical. (that is, play “rules” or “pairing card gambit” rather than chess.)

When numbers one and three—or two and four, etc.—meet with the same score on top board in the last round of a four-round Swiss, something’s gotta give. I have had the third Black out of four myself at least twice. For just this reason, some prefer Swisses with an odd number of rounds.

As others noted, the Toss is still an option allowed in the Rulebook. As usual with options allowed in the Rulebook, we could debate how ‘major’ a variation it is from SOP and how much/what kind of advance notice must be given…but it is an allowed variation. I have used it myself as a TD, though very rarely. (2 or 3 times in club events with small prizes and a mostly relaxed atmosphere)

One down side to a Toss on all applicable boards, down to the lowest score groups, is that a player who shows up on time does not know how to set up the board, or which side of the clock to start, if his opponent is late.

Also, note that some TDs might use the Toss option even if it does not say “Toss” on the pairing sheet. Computer pairing programs can be set to allot colors by random to players with the same score and color history.

A motion to MANDATE last round Toss on all boards where players have the same score and same color history would never pass. A motion to clarify or reinforce what is already allowed as an option in the Rulebook might pass…but why is it needed?

10

I thought the Swiss Gambit was listed in all the opening books. :wink:

11

I doubt whether anybody would attend or not attend a tournament based on whether there is a toss for last-round colors. Thus, it’s a minor variation. It should, however, be announced (or posted) at the site during an early round.

Tossing, or not tossing, is not inherently unfair to either player. If “higher-rated” gets due color, that’s true whether that due color is white or black. So it’s 50-50.

Also, “higher-rated gets due color” is an oversimplification. First of all, it’s higher-ranked, not higher-rated. If two players have different scores, the higher-ranked player is the one with the higher score, not the higher rating.

Moreover, if one player is more strongly due a color than the other, that fact (rather than higher-ranked) would determine color. For example, going into round 5, a player with WBWB is “more” due white than a player with BWWB.

Bill Smythe

12

In a 6-round tournament, suppose only two playes, A and B, are 5-0.

If A’s history were BWBWB, and
if B’s history were BWWBB,

I’d bet that more players in B’s situation would object to tossing for color than would players in A’s situation getting assigned Black.

13

Suppose the colors of two players going into the last round are as follows:

(1) BWBW
(2) WBBW

Both have had an equal number of blacks and whites, and both are due black. The rules state, I believe, that in this situation, you look at “color history” before “rank”. You look at the last point in the color histories where the two players had different colors and that determines the colors. In this situation, the last time the two players had different colors was Round 2. So in this situation player 2 gets White, irrespective of the rank (score/ratings) of the two players.

Steve Immit’s hypothetical was:

(A) BWBWB
(B) BWWBB

Again, both players have had one more Black than White and are due White. White would also alternate colors for them both. But the color history rule is clear. The last time that the two players had different colors was Round 4; and that decides it. Player B gets White. No toss or looking at ratings needed.

The situations where a toss would replace “rank” as the method of deciding colors would be limited to those where the two players have had exactly the same color history in the tournament. Is that when Fred is proposing to use a coin toss? Or, Fred, would you be planning to flip a coin in the above situations as well? If so, I am not sure that the rules variation which permits tosses allows that.

14

This would seem like making a case for “the path of least resistance”. If a director is consistent in making color allocations those objections should be denied.

I used toss up until I started using pairing programs at which point I let the program decide while checking to see if the program got it right. I also had a situation very much like Steve described decades ago but stuck to my assignment of toss because to change it because of a complaining player in my opinion would have been worse. Players also like consistency from the TD. How does the TD defend himself from the charge of “you always did it the other way before and now you’re changing it because my opponent complained”?