Seems like this thread needs to be split up.
Oops. We just doubled the size of the tournament!
Alex Relyea
Thank you. 
Of course, I meant one player due white who gets black, one player due black who gets white, and four players who get their due colors.
Bill Smythe
And here I was thinking that you were combining the two roundsâ pairings and the six games played spanning rounds two and three would include one due black vs due black, one due white vs due white and four due black vs due white.
And if the players with White all win round one?
Note that you are fixating on colors. The reality is that no matter how you pair the first two rounds of a six player tournament, there is a third round pairing that makes a round four pairing impossible. So donât tell me that if you just do normal pairings in round two if all the 1.0âs are due black that things will be fine. Doing potentially highly suboptimal pairings in round two just so you could tell easily what potentially suboptimal pairings in round three would allow you to pair round four isnât a âsolutionâ.
Actually, it would work out that way, too, if all the colors âworkedâ in round 2.
Furthermore, under that same assumption, all the players would get their due colors in round 4, as well.
Bill Smythe
Thatâs definitely true, and you could say that colors are not the ârealâ problem. Itâs just that excessive color alternation is one way (probably the most likely way) to fall into the trap.
So, my advice in the 6-player case was only that, if all colors alternate in round 2, just make sure that in round 3, one player due white gets black, one player due black gets white, and the other four get their due colors.
If white (or black) wins every game in round 1, I would not suggest making all the colors alternate in round 2, even though that would make it easier to spot trouble later, when round 3 is paired. As you said, that âsolutionâ would be suboptimal.
In that case you could use the âhexagonâ method instead, as outlined in (for example), http://www.uschess.org/forums/viewtopic.php?p=201857#p201857.
The hexagon method can always be used to pair round 3 in a 6-player event, whether or not colors alternated in round 2. That way you wonât have to use the Crenshaw-Berger tables, which could be a problem if you donât happen to have your rulebook handy when you pair round 3.
Bill Smythe
This topic is a spin-off from Minimum rating floor in the online rating systems.
Sorry, too difficult to try and unwind whose quotes Iâm unwrappingâŚ
The other take-away was did the Masters learn that weird things may happen if entering an open section? I take the points upthread about doubling the lower rated player against someone without hope of a prize could be the way to go if possible. I understand the Swiss principle of finding the best player at the tournament that Nolan mentioned, but if there are other prizes than first one has to also be sure they arenât disturbed either lest one get anger from the next one down the line. Were there also overall prizes that could be affected, or just the single section? Or that an other-than-natural pairing doesnât redistribute the prizes inequitably (which it shouldnât with one-prize-per-player but maybe there is a way to mess that up tooâŚ)
And, given how the pairings were, would the Masters not have known that they might need to play for a win to succeed? Did any of them have early short draws that they could have played harder on earlier in tournament? (nooo, that never happens in upper level play! 
But may not be the case here.) And while it might not have been a pushover, if the Expert had still won his last game against whatever pairing⌠the Masters would still be upset because they lost out on the money, no?
But then, spinning out a further hypothetical, letâs say the 800 player was pairing into a 2020 and by a most unlikely chance the lower rated player upsets the upper rated player. You havenât disturbed the Masters, touchy lot they can be, but you now have a different Expert who dumped 20 points very irate because the native pairings werenât followed. Or maybe the 800 player isnât happy because he wanted the higher rated pairing because he wasnât playing for money anyway. Like a tiebreak, the potential for inequity exists if things donât go as predicted.
Lower rated players (or their parents entering them) donât necessarily enter open sections to try and win (either money or places) but rather get the best games possible against higher rated players. Players that will not give a Class F the time of day let alone a game. The reality is that for learning purposes it may be entirely possible for the player to get the same educational benefit by playing lower rated players, sure. But if one is relatively sure of being crushed in any event then perhaps the desire is to be crushed by the best player possible. Should that affect the standings⌠Well, it was an open section, yes?
One last question from the peanut gallery, and it may be the most important one I have: If FIDE Dutch pairings are followed, is there any discretion at all? (I donât run nor play FIDE tournaments, I have only understood from others that the Dutch system is completely deterministic and may have misunderstood that.)
The Dutch system (or any other pairing system) has nothing to say about who is eligible to play in a section. The tournamentâs advance publicity provides (or should provide) that information. The pairing rules only tell you how to make the pairings, once eligibility has been decided.
Some say the Dutch system is deterministic, others say it falls a bit short of that. I think recently there was an attempt to make the Dutch system rules a bit more readable, but this may have come at the cost of making the system ever-so-slightly short of deterministic.
Bill Smythe
Thanks for the clarification - I wasnât looking to determine eligibility using the pairing rules. I was wondering how much (if any) discretion a director would have using the system to make a judgement call on the pairings using Dutch rules. Perhaps a different but more precise question to ask would be how FIDE would look upon such a change if a complaint were made about breaking the pairing rules in such a situation versus USCF. My impression (again as someone who doesnât play nor direct FIDE) is that they would be less forgiving of breaking the rules to find a more ethically equitable tournament than the USCF is. I could be very wrong about this as well.
I think it largely depends on whether or not it is a tournament offering norms.
Alex Relyea
Itâs almost deterministic. However, if, e.g., you are using the Dutch system to pair a six player four round tournament, and you run into the round four trap (more likely under Dutch rules than US Chess rules because color is so important), the Dutch rules will fail and the TD would have to use their best judgment how to pair.
The current Dutch system is basically an attempt to document the pairings done by the JaVaFo program, and the newer version of the description is certainly more understandable than the old one. However, Iâve found that following the written rules doesnât produce the same pairings as JaVaFo in quite a few gnarly situations. (Note also that one can theoretically get a program validated by FIDE with some differences from the JaVaFo pairings, so I guess they understand that there is no way to completely document the ârulesâ).
It doesnât surprise me that a system that tries to be deterministic runs into serious problems in specific situations, such as small tournaments. Also, it probably would harrumph if the arbiter wanted to pair two players against each other because both entered late, or because both were on time but their opponents never showed up so they were re-paired against each other, etc. Itâs those little human touches that disappear with determinism.
Bill Smythe
As Alex indicated, whether itâs a norm tournament or not matters. Iâve pretty sure that pairing late entries or re-pairing no-shows is strictly out-of-bounds in a norm tournamentâpairings once posted stand, even if players drop out or if some previous scores were wrong.
I was informed that the âDutchâ rules are not a description of the implementation of JaVaFo, but of âSwiss Masterâ instead.
This topic was automatically closed 730 days after the last reply. New replies are no longer allowed.