Top two groups are as follows:
2.5
2658 BWB
2559 WBW
2
2176 BWB
2080 BWB
2027 xBW
2658 and 2559 have already played. 2080 and 2559 have already played. How do you pair the fourth (and final) round?
Alex Relyea
I’m assuming you are asking in terms of USCF rules instead of FIDE rules.
If you are using 200 for color equalization then either
2658-2027
2080-2559
2176-???
or, preferably per the TD tip but not actually mandated,
2658-2027
2176-2559
2080-???
If your equalization max is less than 149 then the original match-up remains of
2658-2176 (the 2176 gets black for 75% of the tournament)
2080-2559
???-2027
Looking at the 2176-2080 transposition (96 points) doesn’t actually change anything because there would still be one game (board one) with both players due white for equalization. So a further swap with the 2027 is still a 149 point transposition.
The key question is, which player with 2.0 gets dropped into the 1.5 group?
Option 1. If 2027 is dropped, there will be a bad color equalization on one of the top two boards. Not good.
Option 2. If 2080 is dropped, the colors can be made to work on both top boards. Ideal.
But then what happens in the 1.5 group? What if it turns out that option 2 adds a bad color equalization in the 1.5 group? Then nothing has been accomplished. I’m not a fan of simply pushing the bad color down to a lower score group – if a transposition does not decrease the number of bad colors, then don’t transpose.
So: Use option 2 unless it creates an additional bad equalization in the 1.5 group, in which case agonize between options 1 and 2.
Um, Jeff, you overlooked that 2559 has already played 2080.
Bill Smythe
Corrected after Bills comment. 
I’ll blame having spent a long day at a wake.
I’m assuming you are asking in terms of USCF rules instead of FIDE rules.
If you are using 200 (or even 100) for color equalization then
2658-2027
2176-2559
2080-???
If your equalization max is less than 99 then the original modified match-up remains of
2658-2176 (the 2176 gets black for 75% of the tournament)
2027-2559
2080-???
(swap 2080 and 2027 to avoid a rematch since the 2027/2080 swap is only a 53 point difference, the 2080/2176 is a 96 point difference and both are less than the 99 point difference for 2559/2658)
Well, Jeff, it appears from the MSA crosstable for this event that the OP (or his computer) did it our way.
That crosstable does not show colors. But starting from the color info furnished in the original post, and using common sense and basic knowledge of USCF color rules, I was able to deduce (I’m pretty sure) all the colors in that section.
Dropping the 2080 into the 1.5 group did indeed enable perfect colors on the top two boards. But how did it affect colors in the 1.5 group itself? Apparently it helped there too. The 1.5 group consisted of just two players (plus a third who had dropped out and was not to be paired for round 4), one of whom was due white and one due black. The 2080, due white, was paired against the 1.5 due black. This, in turn, dropped the 1.5 due white down to the 1.0 group, where two of the three were due black. Perfect!
If my deductions as to colors are correct, then every player in that section alternated perfectly, start to finish, except for one who ended with WBBW and one with BWWB. It doesn’t get any better than that.
I attribute this degree of perfection to the willingness, on the part of the TD and/or his pairing program, to look ahead (into the following score groups) when deciding which player, in an odd group, should be treated as the odd player. There’s a lesson here! Some TDs would simply drop the lowest-rated player into the next group. In this case that would have resulted in two bad color equalizations, including 3 blacks out of 4 on the top board.
Bill Smythe
Sorry for the delay. I’m always out of commission after running tournaments.
The reason I posted the question in the first place is because SwissSys 8.892 did in fact do
2658-2176
2027-2559
which caused 2176 to appeal (for good reason IMO) and I, after carefully consulting the rulebook and not coming up with any good reason to pair them that way, overruled the computer. FWIW, it was set to the standard 80-200.
Alex Relyea
I wonder if 80/200 only works when staying within a scoregroup for your version of the program. Since I use WinTD instead of SwissSys I can’t really test it.
Aw, shucks. I knew there was a story there somewhere, but I was hoping to hear something a little juicier – like a fistfight between the chief event TD and the chief section TD, after the former overruled the latter.
Another lesson – it’s a good idea for the TD to examine the computer pairings before posting them or showing them to anybody else. This is especially true in (a) the later rounds, (b) the higher score groups, and (c) small tournaments (or small sections). Looks like you had all three of these conspiring against you here.
Bill Smythe
Amen to this! It is also an excellent demonstration of why there should not be an all-fired hurry to post pairings milliseconds after they come out of the printer.
WinTD would also pair them Ivanov-Carter and Vigorito-Schalk.
This is the pairing logic on the computer chosen pairings:
[code]**********Final Pairings
Score Group 2.5
99I EQ LO Ivanov, Alexander(1:2658,W) vs Carter, David E(3:2176,W)
Drop 0.5 Points
149I AL LO Vigorito, David(2:2559,B) vs Schalk, Alan P(5:2027,b)
Drop 0.5 Points
Score Group 2.0
149I OK LO Sage, J Timothy(4:2080,W) vs Welling, Aashish(7:1861,BB)
Drop 0.5 Points
Score Group 1.5
135I OK LO Tang, Arthur(6:1957,W) vs Gaspar, John(9:1955,B)
Drop 0.5 Points
Score Group 1.0
39 OK OK Potorski, Gerald F(8:1994,B) vs Chevalier, Tim(10:1732,w)
Score Group 0.0
0 Albin, Dmitry(11:nnnn,WW) Bye
Total value 1107.253009
0004 Interchange
0001 Wrong Equal
0001 Wrong Alternate
0004 Over Low Limit
0004 Drops for 2.000000
[/code]
and this is on the TD override:
[code]**********Final Pairings
Score Group 2.5
248I OK HI Ivanov, Alexander(1:2658,W) vs Schalk, Alan P(5:2027,b)
Drop 0.5 Points
0 OK OK Vigorito, David(2:2559,B) vs Carter, David E(3:2176,W)
Drop 0.5 Points
Score Group 2.0
149I OK LO Sage, J Timothy(4:2080,W) vs Welling, Aashish(7:1861,BB)
Drop 0.5 Points
Score Group 1.5
135I OK LO Tang, Arthur(6:1957,W) vs Gaspar, John(9:1955,B)
Drop 0.5 Points
Score Group 1.0
39 OK OK Potorski, Gerald F(8:1994,B) vs Chevalier, Tim(10:1732,w)
Score Group 0.0
0 Albin, Dmitry(11:nnnn,WW) Bye
Total value 1108.644861
0003 Interchange
0001 Over High Limit
0002 Over Low Limit
0004 Drops for 2.000000
Best Results Achieved on Pass 1
[/code]
It’s a really close call on the scoring. However, the Ivanov-Schalk pairing grades out as a 248 point drop (2658-2559 + 2176-2027), which is too high for fixing one equalization under the 200-80 rule. This is probably a good indication that the 200-80 rule isn’t a good choice for a 4 round, 12 player section.
I was under the impression that you used either switch to count the points for the 200-80 rules, not both. Am I incorrect?
Alex Relyea
You’re right, Alex: it’s supposed to be the smaller of the two differences, not the sum of the differences.
From the 5th edition rulebook:
You would be correct if you were pairing players in the same score group (a “homogeneous score group”, in FIDE terms). However, in this case, you are pairing “downfloat” players (as the two players with 2.5 have already been paired, you are looking for opponents for each player from the 2.0 score group). In this case, you are supposed to consider only the difference in ratings of the natural opponent and the chosen opponent in the lower score group. The difference between the ratings of the two downfloat players is irrelevant.
Sorry, Bob, but I believe that only applies to players in the same score group, not to “downfloats.” (Your posting appeared while I was composing my reply, apparently.)
Ken, take a look at Example 5 of rule 29E7, on page 160 of the 5th edition rulebook. It’s the same principle: you use the smaller of the two rating differences.
Rule 29D on page 140 says:
So yes, the rules for transposition also apply to switches to improve colors for the odd player.
Admittedly the situation at Alex’s tournament is a little more complicated than Example 5 of rule 29E7 because there are two odd players (downfloats) instead of just one, so let’s take a look at it.
The natural pairings, after adjusting for the fact that 2658 vs. 2559 would be a rematch, are:
2658 BWB (2.5 pts) vs. 2176 BWB (2 pts)
2080 BWB (2 pts) vs. 2559 WBW (2.5 pts)
2027 xBW (2 pts) drops to the next lower score group
The proposed switch to improve colors is:
2658 BWB (2.5 pts) vs. 2027 xBW
2176 BWB (2 pts) vs. 2559 WBW (2.5 pts)
2080 BWB (2 pts) drops to the next lower score group
How should we evaluate this? I think the way we should look at this is first to find an opponent for the 2658, then find an opponent for the 2559. Whichever player is left in the 2 point group becomes the odd player and is paired down.
The first switch is to pair 2658 vs. 2027 instead of 2176. Assuming the rating of the 2027’s natural opponent in the next lower score group is X, the switch should be evaluated as (2176-2027=149) or (2658-X), whichever is smaller. Regardless of what X is, this is less than 200 points, so it’s acceptable since we’re making this switch to equalize colors.
Next we find an opponent for the 2559. The highest rated player remaining in the 2 point group is the 2176, and the colors are O.K., so we pair the 2176 (white) vs. the 2559 (black). No switch is necessary so there’s nothing to evaluate.
So actually I agree with Ken that the rating difference between the 2658 and the 2559 is irrelevant in this example.
The algorithm for evaluating differences between “natural” and “proposed” pairings is designed to give an objective value that can handle even multiple drop situations like this. The natural pairings here would have the two 2.5’s play each. Of course, they can’t. One could try to figure out the most natural pairings that don’t run result in duplicates, which would here be Ivanov-Sage and Vigorito-Carter, and evaluate color corrections from that. That would require (in effect) doing pairings twice and in practice there would be multiple choices for that first set of “second best” pairings.
If there is a drop from the 2.5’s to the 2.0, the pivot pairing is the bottom 2.5 playing the top 2.0 (2559 vs 2176). All the 2.5-2.0 pairings are evaluated based upon the distances from that. Ivanov playing the the high-rated 2176 would be 2658-2559=99 since you’re switching only one of the players from the pivot. But Ivanov playing the low-rated 2027 is a change on both sides, thus the need to sum the distances. Inside a single score group, you would always be able to look at a change as a switch on just one side, but not when the players are in two score groups.
This is the same value to the swap that you would get if there had been three 2.5’s and you were considering dropping the highest ranked 2.5 to play the lowest ranked 2.0 to fix colors, which most TD’s probably wouldn’t do under USCF 200-80 rules.
That’s not what the 5th edition rulebook says. You’re supposed to look at the smaller of the differences, not sum the differences. An illustration of this is Example 5 on page 160 (rule 29E7). True, Example 5 isn’t quite the same as the situation at Alex Relyea’s tournament, but it does show the principle of looking at the smaller of the differences even when players are in different score groups. Can you cite any rule which talks about rating differences being added when evaluating a transposition?
This makes no sense because both the 2658 (Ivanov) and the 2559 are odd players. According to Example 2 on page 140 (rule 29D, The Odd Player), the natural pairing is 2658 vs. 2176, 2559 vs. 2080, so for Ivanov to play the 2176 is a rating difference of 0. In effect there are two pivots in this example, not just one.
Ivanov is already an odd man, and the question is which player with 2 points he should play. Why would be it be a “change on both sides” for him to play the 2027 instead of the 2176?
Consider the case that I mentioned. Suppose that you have three 2.5’s and need to pick a down floater. How do you evaluate the size of a switch that has the highest ranked 2.5 playing the lowest ranked 2.0? Whether it’s explicit in the rule book or not, the most obvious value is to sum the gaps since you’re changing both the downfloater and the upfloater.