| |
BGonline.org Forums
Madison - Thank you..yes indeed :)
Posted By: neilkaz In Response To: Madison - Thank you..yes indeed :) (Matt Cohn-Geier)
Date: Wednesday, 3 September 2008, at 10:31 p.m.
MCG wrote: I absolutely agree with that, and would have been very upset if I had been informed at 7-2 that I was playing for 2nd place. If you're going to run a straight Swiss, you need to announce that the players with 2 losses may be out before the event starts.
Neil agrees completely and also wouldn't have been happy: As I continue to give thought to what happened and strive really hard to avoid late byes, I'll again appologize for not mentioning all the good things that would have occurred had we played a 10th (just the 4 of us) round. We're all fair people. I for one, would have been willing to proclaim Steve winner if he went 10-0. So lets say he'd played Tim and beaten Tim. Tim is 7-3 and out. You beat me..I am 7-3 and out and share 3/4 with Tim. You're 8-2 and get 2nd place. I'd accept that as fair and the T ends. Thus if someone ever goes 10-0 they should be given the title in this (modified) Swiss.
But lets say Steve lost and was 9-1. Tim and you are now 8-2 and I am 7-3 and can either be awarded 4th place (4,2,1,0.5) scale or go to the conso and have to win 2 matches to get into the money as I did. You and Tim play a match and the winner at 9-2 plays Steve at 9-1 for the title in a single match.
That 10th round idea seems fairest to me and it is most like a real 100% Swiss. In a true 100% Swiss if Steve was 10-2 the lone 8-2 guy has to beat him thrice and therefore Steve's 87.5% to win the T from 10-0. In a 100% Swiss had Steve gone 9-1 and there were two 8-2's there's no fair way to resolve since the bye is huge as three players remain. Having 8-2's play down to a 9-2 and then beat Steve twice doesn't seem proper since after the first win a 10-2 is playing a 9-2 one match for the title, but having the winner of the 8-2's play just one match vs a 9-1 also isn't perfect. A coin could be flipped as to whether it would be one or two finals matches and that would make Steve 62.5% to win from 9-1 which when averaged with 87.5% in a true 100% Swiss from 10-0 makes Steve 75% to win at 9-0. Which is just what he'd have been under my proposed (ashamed I didn't think this up Sun night) rd 10 method where if Steve wins he's got the title and if he loses rd 10 he's in the finals and 50% for the title.
So in summary of my ramblings so far...it seems reasonable to me that after rd 9, Steve at 9-0 should be 75% to win the T, noting three 7-2's remain. This seems far more fair to Steve than being 50% as he actually was and having to play a 9-2 one match for the title. As resulted Tim at 10-2 won the event and Steve at 9-1 was 2nd. A huge congrats for Tim, but as has happened in many Swiss playoffs this doesn't seem completely fair to me.
I do feel strongly that 10-0 should win the event, period. This would have simplified things a lot and no silly 1 game RR for bye etc. I have no problems with an unspecified number of rounds for the hoped for almost 100% Swiss-like event and note that it will only be 10 in the unlikely event someone wins 10 in a row, and almost always will be only 12 rounds otherwise.
Rambling still..lets say that Brad beat Steve in rd 9. Now Steve is 8-1 and Neil, Brad, Tim, Matt are 7-2. The best way to handle that, IMHO would be to put Steve into the finals, making him 50% to win the T, and for the remaining 4's 7-2 dudes to play down to 9-2 to play Steve one match. Is it fair to have a 9-2 equal to a 8-1 ? Well it's a hell of a lot fairer than having a 9-2 equal to a 9-0 as happened. Lets say the 9-2 wins the final..he's 10-2 and the loser is 8-2 and 10-2 is lots better than 8-2 so he deserves first place. Now lets say 8-1 player wins..he's 9-1 and that lots better than 9-3.
OK so we can begin to see a few things from my proposed method (note that if 80 players show up some year, we may need another round to clarify) but for now..we have for my propose 10 rd full Swiss.
Someone who goes 9-0 is 75% to win the T Someone who goes 10-0 is 100% to win the T
Someone who goes 9-1 is 50% to win the T. He should be 75% to win the T if there's only one 8-2 left.
Lets say we had a bigger event and there was a 9-1 and 3 8-2's remaining. Clearly all four are getting a prize for time reasons of undesirabilty of feeding down to conso that late. Anyhow..9-1 plays an 8-2 and if he wins we have a 10-1 remaining with a 9-2 who then has to beat the 10-1 twice..and in true Swiss format either a 11-1 wins or two 10-2's play one match for the title. Note that this takes 13 rounds but we'd only possibly have a 9-1 and three 8-2's with a considerably bigger turnout and all later matches could be kept 9 pts if needed for time constraints. If the 9-1 loses we have 3 9-2's who then, using the fair adjustment scheme (similar to Gregg's idea) playoff for either 1st,2nd,3rd or 1st with 2nd/3rd tieing. The 4th dude who's 8-3 gets 4th place and his take is not affected by what happens to the remaining three. The ABT pts are also paid to the remaining three based on the fair adjustment formula.
Lets say, we'd had four 7-2's (I expect that may happen sometime) well..then it is playoff time..duh :). Same thing if we'd had somehow eight 6-2's..playoff time. With a larger event we might get four 8-2's someday.
So in summary (finally Neil's shutting up..don't bet on it) I think I've shown ways that we can almost always handle almost anything can occur. Unless a finals can be set up fairly and without byes up to a 10th round of the Swiss and perhaps an 11th(bigger event) should be played. 10-0 wins the event period ! 9-1 gets to the finals almost certainly. If a lone 8-2 remains to play a 9-1 or a lone 9-2 remains to play a 10-1 the player with one less win and one more loss must beat the leader twice. Again there's the issue of what to do if a 9-1 meets a 9-2, but in the case of a 9-0 winning he's 10-0 and given the title, and if he loses he goes to the finals and when a 9-2 meets him they only play 1 match. In the case of an 8-1 becoming a lone 9-1, again he goes to the finals. In keeping with what I'm proposing, if an 8-2 meets him in the finals, the 8-2 must win twice but if a 9-2 meets him, there's only 1 match. Anyhow..by continuing the Swiss to a 10th and occasionally 11th round and by limiting those playing those late rounds, if possible to those with two of fewer losses, the issue of late byes is minimized. The conso does need to be set up after 9 or 10 rounds or a bye into the money may happen there. Once again, anyone who got a bye earlier in the Swiss is obviously ineligible for a late bye.
OK in closing..what do we do, if we, depending on the size of the event have three players remaining and equal at 7-2 or 8-2 ? Well we use my proposed adjustment to Gregg's good idea to compensate those who don't get the bye. We also pay out the ABT points based on that adjustment and if we do that, then there's absolutely no gain (except being more likely to take home a trophy) from getting that late bye.
I am very convinced that it's not hard to solve almost any issues resulting from this Swiss Tourney format while keeping the event almost entirely Swiss. Byes will happen in the conso, but hopefully not into money rounds. Byes will be very unlikely into the money in the main. In the case of three remaining equal players, use a proposed equity settlement for those who don't get the bye. If we have 5 equal remaining players, run an extra round of Swiss prior to conso. With 6 remaining equal players, play down to 3 etc. Once again, unless we have a considerably larger field, we are likely to need only 12 rounds, just like this past weekend.
.. neilkaz ..
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.