Discussion in 'Archive' started by NoPoke, Jan 25, 2004.

8 league13 468 60
  1. NoPoke

    NoPoke Active Member

    So what is the best way to resolve the following...

    based upon match record

    a 28 year old is to be paired against a 9 year old ( the 9 year old has the better match record ) and a weaker 11-14 player gets a bye.

    the alternative is to give the 28 year old a bye and pair the 9 year old against a different weaker 11-14 player
    the 11-14 player has the same match record as the 28 year old but has already received a bye.


    help please. Neither option seems ideal.
  2. ScythKing

    ScythKing Member

    Yer right - none of the options really seems "right" and Gawd knows the parents usually have a cow when their tyke gets paired with someone 3 times their age - so I use option 2 - a different 11-14 weaker player and give the older player the bye - he could probably use it anyway if it came down to this LOL.
  3. mysterioustrainer

    mysterioustrainer New Member

    I would propose the 11-14 year old player get the bye since the 9 year old is doing so good, and the 28 year old already has earned a bye.
  4. killa_project

    killa_project New Member

    give the 9 year old the bye and have the 11-14 battle the 15+
  5. meganium45

    meganium45 Active Member

    Unfortunately, you have to go with the older player playing the younger player.

    Unless you have a situation where the younger players should be matched to play for a title, the software dictates that a match should be had, then the players should play.

    It happens all the time at major events. It is the only fair way to run an "age-modified-swiss" event where the participants can possibly be playing against each other.

    The ONLY option is to do what Steve does in Colorado, which is actually hold the event as 3 separate tourneys, which if we get the OK to do so is THE way I would prefer to run my City Championships, so each age division can play to a clear winner.

    I find it an atrocity that, like in s R/\s event, where the 10 and under age division can be decided by the 2 top players, with equal records who had not played each other that day, out of the same age division having a title decided by them playing Different opponents from Different age groups. Neither player felt it was right, but it was what the computer said to do. On that circumstance, I would have overridden the computer's wisdom. neither player felt that the result was just, and both would have felt better if they had faced off to determine the title head to head.

    given that the manual pairing input is not working on the software at this point, good luck.

    Given that it took me over 2 hours to manually enter a 54 player event for a prerelease, really good luck.

    My 2 cents, see you at my events

  6. DaytonGymLeader

    DaytonGymLeader New Member

    Manual Pairings works fine. You just have to know how to massage the software. We've spent a couple of months beating the crap out of it and getting it to tell us its secrets. It wasn't pretty. Many Bothans died horrible, messy, minty fresh deaths. TC should be posting the guide here soon.

    Personally, there's more info needed. Exact records, have any of these players faced each other earlier in the event, etc, etc. We're starting to get into territory that we went through with WotC as well here. Quite a few of us lobbied for separate events for each age group under WotC, looks like it may be time to again?
  7. meganium45

    meganium45 Active Member

    Ah, Steve, I meant manually inputing your results match by match.

    The website is not letting me do that anymore. IF you have had luck with it recently, let me know.

  8. SteveP

    SteveP Active Member

    meganium45, the manual online reporting was finally up last thursday and sunday when I reported our last 3 three tournaments.

    NoPoke, without seeing the entire player standings and ages, it's hard to tell what would be proper in your situation. If those player are near the bottom of the standings, then giving one of them the bye won't be too much of a problem. Anyway, here's my take on the age-modified system:

    Regarding age-modified pairing, I reverse-engineered the WOTC DCI Reporter algorithm for this and came up with the following (without going into the details of how complex and recursive this algorithm is):

    Pair players first by record (including any tiebreakers)
    Pair players second by age
    Pair players third randomly

    but, of course if the players already played, you might end up with an unusual pairing such as the top player playing someone 5th in the standings, or an adult playing a child (10-under) with matching records.

    As far as giving an adult the bye, that's VERY unusual. That adult would have to have the worst record with no one younger having the same or worse record, or the younger player already had a bye.

    Players at or near the top should NEVER, EVER get a bye, even if the TO/HJ is doing pairings manually. Any TO/HJ doing this would be SERIOUSLY violating the purpose of the age-modified pairings system.

    So, the only PROPER thing to do is to separate the age groups and run separate tournaments. If that's not possible, you're stuck with the age-modified system where it's very common for younger, decent players to play adults. That's just how it works. Anyone the FORCES pairings that are contrary to the age-modified system, especially at such events as the CC, IMO are in SERIOUS violation of running IMPROPER tournaments.

    Finally, here's something that I occassionally do to re-adjust the pairings. There is a bit of randomness involved in age-modified pairings, especially in the early rounds. If you do the pairings and see something you want to try to avoid (such as a 9-year old playing an adult, or the two best players playing each other in the first round), then delete the round and re-pair. The randomness might create pairings that are more desireable.
    Last edited: Jan 26, 2004
  9. NoPoke

    NoPoke Active Member

    The tournament software went bang again so the whole thing was run manually.
    Increases the stress and workload on the TO!.
  10. meganium45

    meganium45 Active Member

    Steve, I agreed with you completely until...

    you mentioned a situation where the two best players are playing each other in the first round, then deleting the pairings and repair....

    this artificially inflates and deflates the difficulty of the tourney for several players.

    1. For the players who get to avoid playing each other in the first round - makes their first round match easier.

    2. For the rest of the players who play in the first round, makes their first round matches more difficult.

    I find in invigorating to have a great player, by the luck of the draw, lose round 1, and have to fight their way through the losers bracket. It is good exprience for any player, preparing for a large event to know how to fight up from the bottom. You of all people should know that from Origins....

    Just my 2 cents worth, but as you know, I have been wrong before.....

    Talk to you soon

  11. NoPoke

    NoPoke Active Member

    having two strong players meet on the first round is always problematic in a tournament with a small number of rounds. The loosing player will have an easier ride up through the rest of the rounds, often able to sneak the tournament win at the final round if you run 7 rounds. Unless there are a lot of entrants it is probable that all players will have at least one loss by the time round 7 is finished.
  12. SteveP

    SteveP Active Member

    meganium45, the arguments on this can go both ways. I remember a few years ago when the DCI Reporter software actually allowed you to include a player's rating so that the pairing algorithm also took that into consideration. How it handled that rating when doing the pairings, I'm not sure, but I imagine it was probably used to avoid the first-round situation I explained above.

    Nevertheless, I'll concede that the practice of re-pairing ANY round can give the appearance of being biased. It lessens the randomness if you're trying to force/avoid certain pairings.

    Finally, I tend to agree more with NoPoke on this issue of first-round, top player pairings. Later-round losses tend to be less detrimental because you're loosing to a better opponent. Plus, your victories tend to increase your opponent-difficulty tiebreaker when you're playing against unbeaten opponents. I'd bet my paycheck that if your only lose came in the final round, your ranking would be higher than if that lose came in the first round. That's just how the swiss system works. Knowing that, the argument can easily be made to give top players an easier opponent in the initial round. JMO.
    Last edited: Jan 26, 2004

Share This Page