players with byes had to play each other 2nd round

Discussion in 'TCG News & Gossip Discussion' started by TrEkIeV, Apr 19, 2004.

8 league13 468 60
Thread Status:
Not open for further replies.
  1. TrEkIeV

    TrEkIeV New Member

    at my state championships, players who had received 1st round byes from winning CC's all ended up playing each other in the 2nd round rather than another random player. I know the reason they were paired against each other was because they had the same record, but isn't this a disadvantage?
     
  2. yoshi1001

    yoshi1001 Active Member

    Probably has to do with how the software works. Anyone got some insight?
     
  3. SteveP

    SteveP Active Member

    Please elaborate. What the TO using TMS?

    The pairings should've been random. It's unlikely that ALL the CC Champs were paired against each other, but it's obviously a possibility that can't be ruled out. Afterall, the CC Champs have the same record as those who played and won in round 1.

    I'm just curious if you're exaterating here or it really did happen like you said.
     
  4. DaytonGymLeader

    DaytonGymLeader New Member

    I agree. Need more meat to figure this problem out. Were all these folks in the asme age group? Tie breakers? etc, etc, etc.
     
  5. SteveP

    SteveP Active Member

    I'm curious. Does anyone know if TMS or DCI Reporter uses tiebreakers during automatic pairing? It's been my understanding that tiebreakers are ONLY used for standings, not for doing pairings. I thought that only match SCORE (and AGE CATEGORY for age-modified) were used for pairings. If those factors are equal, then the pairing is random (no by alphabet as was the problem in an earlier version of TMS).
     
    Last edited: Apr 19, 2004
  6. TrEkIeV

    TrEkIeV New Member

    i'm 85% sure that all the players that got first round byes had to face each other second round. the PTO said he was using the newest version of TMS.
     
  7. SteveP

    SteveP Active Member

    I just tested this on TMS version 1.11 with 4 awarded byes and 12 players (all in the same age-category). In 10 attempts to pair the second round, only once did it pair the 4 first-round bye players against each other. So, I think it was just random luck that this happened at your tournament.
     
  8. WeileMom

    WeileMom New Member

    We were using the latest version of TMS. It definitely still has a lot of bugs- we had to restart 3 times. I did notice that each time we had to restart that since we had an odd number of people the person who ended up with a bye was a different person each time- therefore I would say the software selection was random.
    I didn't notice what Trekiev did, though there is no reason to doubt him. My guess is that it was purely coincidental since we had so many people there were a lot with the same record going into the second round.
    However, now that I think about it I think the software does do something strange with those receiving first round byes. I remember noticing once that those who had the first round byes, their records showed them playing one less game. So, instead of having a record of 1-0-0, they had a record of 0-0-0, but had 3 points awarded. I'll have to pay closer attention next time to see if that's true. Or Karl, who has more experience with the software, may be able to answer the question.
    The only pairings that weren't controlled by the software were the Top 8.
    ---Nicole
    WeileMom
     
  9. WeileMom

    WeileMom New Member

    SteveP- did you notice what the record was for the people you gave byes to? Did it show up as a win? Or did they just start with 0-0-0 and 3 points?
    ---Nicole
    WeileMom
     
  10. NoPoke

    NoPoke Active Member

    0-0-0 and three points seems like a reasonable way of discribing a player with an awarded bye. Unfortunately I haven't the foggiest idea of how TMS calculates its tie-breakers. I've tried some experimental 8 player tournaments and after a couple of rounds the tie-breaker percentages are completely beyond me inTMS.

    From memory reporter does not use tie-breakers when pairing during the swiss rounds but does use tiebreakers when performing the pairings for the playoffs.

    Lets see we have B players with awarded byes and R players with one win. What is the probability of all the Bs being paired? Another way of looking at this problem is that you have a deck of playing cards with B BLACK and R RED, what is the probability of all the BLACK cards being adjacent?

    [Hint this is very similar to the muligan hand problem ]
     
  11. SteveP

    SteveP Active Member

    Interesting question regarding whether Byes show up on the W-L-T record. In my personal tournament software, they show up as Wins. But, my tiebreakers ignore them, like DCI Reporter also ignores Byes.

    My players would complain if a Bye didn't show up in their W-L-T record, even though in my software, that field is really just informational. The Score field is the REAL indicator of a player's record in the software.
     
  12. old man

    old man New Member

    While I can't say for others I do know that in the 2nd round my son had to play against another player who had a bye in the first round and they are in 2 different age groups.
     
    Last edited: Apr 20, 2004
  13. DARKGeNGaR094

    DARKGeNGaR094 New Member

    TMS also sometimes seems to give a bye to 1 person, but give the win and 3 pts to another person. (in a 5 round tourney, the first would have 4 rounds to his/her name and the latter would have 6)
     
  14. mewsmom

    mewsmom New Member

    We thought this was odd as well. You'd think they would have been paired up in their own age group. But the 15+ buys were all matched so guess he had to go to 11-14. It was the tb/percentages that paired all the Byes together. They all had 3 pts and the same tb/rankings.

    As for the details, the byes were awarded a win but their TB ratings/percentages were all the same and very low and in the end we FEEL this kept my daughter from top 4. Had she not taken her bye and had the same record going into top she would have made top 4.

    So tell me what's the reasoning behind taking a bye if it is only going to hurt your percentages??

     
    Last edited: Apr 20, 2004
  15. WeileMom

    WeileMom New Member

    You know, this could have HUGE implications if the TMS is lowering percentages of those with byes. A lot of T8 and T4 players were determined on tie breakers, with only a couple of percentage points difference.
    Interesting.
    ---Nicole
    WeileMom
     
  16. NoPoke

    NoPoke Active Member

    Last edited: Apr 20, 2004
  17. WeileMom

    WeileMom New Member

    Thanks, NoPoke, I hadn't read that thread.
    Obviously the TMS software wasn't updated to give the .333 % for a bye.
    Have any TO's made PUI aware of this?
    ---Nicole
    WeileMom
     
  18. SteveP

    SteveP Active Member

    TMS should NOT give the minimum 33% for a Bye match. Byes should be ignored when calculating tiebreakers.

    However, the minimum tiebreakers for opponent toughness (Opponent Match-Win Pecent and Opponent Game-Win Percent) SHOULD have a total minimum calculation of 33%.

    For everyone's information, the 33% minimum tiebreaker applies to the accumulated end-result of the tiebreaker calculation, NOT to each match. So, if the tiebreaker for the toughness of ALL your opponents is less than 33%, you should get credit for the minimum 33%. Obviously, the end-of-round-1 tiebreaker (Opponent Match-Win Percent) should be either 33% or 100%.

    FYI, TMS ver. 1.11 does NOT have the 33% minimum tiebreaker yet. Also, I haven't tested it to see if it uses the tiebreakers to do pairings, which would be WRONG if it did. Pairings should be based on 1) Score, 2) Age (for age-modified only), and 3) Randomness.

    Also, in version 1.11 of TMS that I tested, I remember that the players with first-round Byes ALL had the same tiebreaker as those who played and won their first-round game. Once again, I think this is nothing more than a random incident. Without the tournament file, we can't determine otherwise.
     
    Last edited: Apr 20, 2004
Thread Status:
Not open for further replies.

Share This Page