I recently read of a store doing using the double elimination system in place of Swiss. Whilst the scenario was FNM--poor choice, IMHO--and the user framed it as a negative, I can't help thinking that using the double elimination system would actually hold a lot of benefits over Swiss (especially for larger events).
Problems that are lessened or removed in DE:
1. Dropping. Players who drop sometimes fail to report this. DE drops players automatically, saving time and confusion.
2. Pair downs. Whilst these would still exist, they'd be much less likely.
3. Intentionally drawing into the top. This would be less prevalent when there's no "loser's bracket".
Personally, I'd love to see a major event try out this system. Thoughts?
Private Mod Note
():
Rollback Post to RevisionRollBack
I'm officially proposing we retire the word "insane" from the MtG vocabulary.
"The best way to be different is to be better" - Gene Muir
Swiss, when used with a T8 bracket, effectively IS double elimination.
Oh, I realize. Heck, this article lays it out as such (from a mathematical perspective). Still, it's not truly DE since the people with losing records continue to play through the whole event.
What I'm talking about is true, cut-throat DE. Not fit for FNM, but fine for events with hundreds or thousands of participants. It could be tested out at a GP, for example.
Top Eight bracket. In larger tournaments, there will typically be a series of single elimination games played between the eight players with the best records after X number of Swiss rounds. For example, an event with 30+ participants would likely lead to a T8 bracket near the end.
The main problem with Double Elimination is that you're extending the length of the tournament even more. I do like the system (it's used in many fighting game tournaments) but for Magic not as much.
A disadvantage compared to the single-elimination format is that at least twice the number of matches have to be conducted.
Time is a serious factor in competitive MTG. Time is the reason why so many draws exist/has to to exist -- you can't go "well, you have to play until someone wins/loses, draws aren't allowed" . Doubling the time required to complete a tournament makes the problem worse, not better.
I believe double elimination is better at coming up with the better player. However, there simply isn't enough time. As it is, even a medium sized tournament has to cut short the top 8 and everyone in the top 8 just splits it. GPs are two days.
This can probably be adopted for something like worlds, where people fly in anyway and so they can play for 4+ days straight. It is highly impractical for anything else. And even then, it stops being a matter of skill, but endurance.
Time is a serious factor in competitive MTG. Time is the reason why so many draws exist/has to to exist -- you can't go "well, you have to play until someone wins/loses, draws aren't allowed". Doubling the time required to complete a tournament makes the problem worse, not better.
I don't really know how to do the math, but i'm pretty sure double elim is faster than 12 rounds of swiss with top 8.
Oh, I realize. Heck, this article lays it out as such (from a mathematical perspective). Still, it's not truly DE since the people with losing records continue to play through the whole event.
What I'm talking about is true, cut-throat DE. Not fit for FNM, but fine for events with hundreds or thousands of participants. It could be tested out at a GP, for example.
Top Eight bracket. In larger tournaments, there will typically be a series of single elimination games played between the eight players with the best records after X number of Swiss rounds. For example, an event with 30+ participants would likely lead to a T8 bracket near the end.
Horrible idea. Imagine a new player goes to his first GP, flies a decent distance, in fact, loses his first two rounds, and instead of being able to play the remaining 7 rounds for the fun of it and getting planewalker points, he is simply eliminated and told "nope, you're done, sorry".
I don't really know how to do the math, but i'm pretty sure double elim is faster than 12 rounds of swiss with top 8.
the formula for double elimination is
# of games required = (# of players x 2) - 2
or +1 additional game at the end. so 8 players = 14 or 15 games needed to get a winner, 16 players = 30 or 31. i don't know how swiss works so someone can chime in with that to compare.
edit: # of games is worthless, we want number of rounds to compare.
8 players = 5 or 6 rounds
16 players = 6 or 7 rounds
I'm not really sure how this would work out. Let's say that there are 11 players left in an event. Is someone getting a bye this late in the event? And then when that does happen, we go down to 6 players, then 3. When we're at 3, is one player just getting a bye right to the finals?
I'm not really sure how this would work out. Let's say that there are 11 players left in an event. Is someone getting a bye this late in the event? And then when that does happen, we go down to 6 players, then 3. When we're at 3, is one player just getting a bye right to the finals?
There would be enough byes in round 1 to ensure the number of players in the winners bracket in round 2 is a power of 2 so that there will be no need for further winner's bracket byes.
Private Mod Note
():
Rollback Post to RevisionRollBack
Hey all... I'm retired, not dead. Check out what I'm doing these days (and beg me to come back if you want):
The numbers would be pretty reasonable in terms of length comparing a 3 or 4 round swiss + top 8 vs a Double elimination for any tournament up to 128 players. For an event like a GP they already do 8 rounds on the first day alone and you can support 512 players on a double elimination that runs 10 rounds plus a final. Bit of a stretch for one day, and also not really kosher to run it for two days only expecting a handful of players to need to be there for the second day. But this still runs into "Yeah but I want to play the game." Magic at a tournament level is something of a sport, but its also a game that people come to play and have fun without any serious aspiration of winning the event. Cutting out half of the people from the event because they lost their first two games isn't good for that.
The real equation is that a casual thing like FNM isn't really about getting a winner at all. Its an event that has prize support but is mainly for people to come, play magic for a few hours, hang out, trade, let the store sell product etc... you don't do that by having half your field dropped in two games. You do that by having everyone play 3 or 4 rounds and then having a top 8 cut if the field was big enough to warrant it.
How would this work with draws? In a large magic tournament I imagine at least 1 match will go to time each round and if not there are still actual drawn games where both players die simultaneously.
set a time limit for first day, you have to win X matches to get to day 2. You are eliminated as soon as you get 2 loses.
As soon as you win you go to the report station with the match slip. You will be paired up against another player, given the match slip and the table number. The standings won't matter because as soon as you lost twice you are out of the tournament already.
No draws for first day what so ever. If you fail to win X matches when the given time comes, you don't make day 2.
Of course, this will only work for first day.
Private Mod Note
():
Rollback Post to RevisionRollBack
Want to have more creativity with your food?
Wanna hear what I think about restaurants?
Wouldn't a change to DE prompt places that host events to host side events much more regularly (as you would have large numbers of players getting dropped from the event) thus causing people who want more magic to play to have well, more magic to play?
set a time limit for first day, you have to win X matches to get to day 2. You are eliminated as soon as you get 2 loses.
As soon as you win you go to the report station with the match slip. You will be paired up against another player, given the match slip and the table number. The standings won't matter because as soon as you lost twice you are out of the tournament already.
No draws for first day what so ever. If you fail to win X matches when the given time comes, you don't make day 2.
Of course, this will only work for first day.
"no draws" is impossible. Earthquake for 50, what happens? Also, if it were possible to institute "no draws", then what the point of changing from swiss?
As for "you finished, move to next available opponent" Years (1998? 97?) ago, I joined a large tournament with something similar to this.
What happened was several players just played 1 game, then turned in their slip, with the agreement that who won the first game also won the second.
By the time I finished 3 games and realized this, some players were already on their 8th or 9th. The time was called, I lost 1 game, won 6, but there were players who won 15+ matches. Landed 30th place or something.
Also, "lose 2 games, you go home" doesn't address the problem mentioned by previous posters.
Finally, if you institute a "win X to go to day 2", people who already won X will simply stop playing... which _will_ end up with players automatically locked out because they no longer have any opponents left, even if there is time left or they have no losses.
DE works only in games where there aren't commonly draws.
Taking a sporting example, DE could work in Australian Rules football (where scores are usually around 90 a side making ties uncommon) but not in soccer (where a good quarter of matches between teams of comparable calibre end as draws).
Actual draws in Magic are rare but for time reasons they are common, and stalling is a real issue as a result.
The only solution I can think of is to have a system where you are playing DE, but in the case of a timed-out match you have a much more ruthless way of determining the winner - either via an in-game sudden death scenario (as the tournament rules allow for in the rare case of a single elim scenario where time runs out) or via an out of game system (at the end of the time period, if the game is undecided, it's decided by a coin flip).
The 'being bored due to early elimination' issue isn't a problem as DE eliminates people in big groups and in predictable numbers, so side events would be easy to organise. If you start with 512 players, DE will eliminate 128 after round 2 and a further 128 after round 3 and a further 96 after round 4 - all in big groups.
"no draws" is impossible. Earthquake for 50, what happens? Also, if it were possible to institute "no draws", then what the point of changing from swiss?
As for "you finished, move to next available opponent" Years (and years) ago, I joined a large tournament with something similar to this.
What happened was several players just played 1 game, then moved to the next, with the agreement that who won the first game also won the second.
By the time I finished 3 games and realized this, some players were already on their 8th or 9th. The time was called, I lost 1 game, won 6, but there were players who won 15+ games. Landed 30th place or something.
I assume there would be a match limit number. As in you get 8 rounds total to work with and need at least 6 wins to make day 2.
The no draws in this scenario is somewhat easier as the magic tournament structure already does this. Each match is to the best of two which in a match without a time limit like these would would just mean you go to game 3,4,5, 6 as necessary.
Private Mod Note
():
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
Problems that are lessened or removed in DE:
1. Dropping. Players who drop sometimes fail to report this. DE drops players automatically, saving time and confusion.
2. Pair downs. Whilst these would still exist, they'd be much less likely.
3. Intentionally drawing into the top. This would be less prevalent when there's no "loser's bracket".
Personally, I'd love to see a major event try out this system. Thoughts?
I'm officially proposing we retire the word "insane" from the MtG vocabulary.
"The best way to be different is to be better" - Gene Muir
Cubes:
Modern Banlist Cube
Monocolor Budget Cube
Current post- Grand Prix KC Modern Postmortem (7/7/13)
Noah Weil on scouting, an attorney from Seattle with 20 Pro Tour appearances.
Oh, I realize. Heck, this article lays it out as such (from a mathematical perspective). Still, it's not truly DE since the people with losing records continue to play through the whole event.
What I'm talking about is true, cut-throat DE. Not fit for FNM, but fine for events with hundreds or thousands of participants. It could be tested out at a GP, for example.
Top Eight bracket. In larger tournaments, there will typically be a series of single elimination games played between the eight players with the best records after X number of Swiss rounds. For example, an event with 30+ participants would likely lead to a T8 bracket near the end.
I'm officially proposing we retire the word "insane" from the MtG vocabulary.
"The best way to be different is to be better" - Gene Muir
Cubes:
Modern Banlist Cube
Monocolor Budget Cube
A Top 8 bracket.
The main problem with Double Elimination is that you're extending the length of the tournament even more. I do like the system (it's used in many fighting game tournaments) but for Magic not as much.
Time is a serious factor in competitive MTG. Time is the reason why so many draws exist/has to to exist -- you can't go "well, you have to play until someone wins/loses, draws aren't allowed" . Doubling the time required to complete a tournament makes the problem worse, not better.
I believe double elimination is better at coming up with the better player. However, there simply isn't enough time. As it is, even a medium sized tournament has to cut short the top 8 and everyone in the top 8 just splits it. GPs are two days.
This can probably be adopted for something like worlds, where people fly in anyway and so they can play for 4+ days straight. It is highly impractical for anything else. And even then, it stops being a matter of skill, but endurance.
"Sometimes, the situation is outracing a threat, sometimes it's ignoring it, and sometimes it involves sideboarding in 4x Hope//Pray." --Doug Linn
I don't really know how to do the math, but i'm pretty sure double elim is faster than 12 rounds of swiss with top 8.
Horrible idea. Imagine a new player goes to his first GP, flies a decent distance, in fact, loses his first two rounds, and instead of being able to play the remaining 7 rounds for the fun of it and getting planewalker points, he is simply eliminated and told "nope, you're done, sorry".
What a great experience.
The math is in the page.
2n-1 or 2n-2, where n is the number of competitors.
12 rounds means paying 12 times.
in a 36 man tournament, you'd have to play 70 or 71 games per person.
500 man GP? Yeah.
"Sometimes, the situation is outracing a threat, sometimes it's ignoring it, and sometimes it involves sideboarding in 4x Hope//Pray." --Doug Linn
the formula for double elimination is
# of games required = (# of players x 2) - 2
or +1 additional game at the end. so 8 players = 14 or 15 games needed to get a winner, 16 players = 30 or 31. i don't know how swiss works so someone can chime in with that to compare.
edit: # of games is worthless, we want number of rounds to compare.
8 players = 5 or 6 rounds
16 players = 6 or 7 rounds
You can find me on MTGO. My username is gereffi.
There would be enough byes in round 1 to ensure the number of players in the winners bracket in round 2 is a power of 2 so that there will be no need for further winner's bracket byes.
https://twitch.tv/annorax10 (classic retro speedruns & occasional MTGO/MTGA screwaround streams)
https://twitch.tv/SwiftorCasino (yes, my team and I run live dealer games for the baldman using his channel points as chips)
The real equation is that a casual thing like FNM isn't really about getting a winner at all. Its an event that has prize support but is mainly for people to come, play magic for a few hours, hang out, trade, let the store sell product etc... you don't do that by having half your field dropped in two games. You do that by having everyone play 3 or 4 rounds and then having a top 8 cut if the field was big enough to warrant it.
Hoping for a cure, or at least an outbreak.
Level 1 Judge (yay)
set a time limit for first day, you have to win X matches to get to day 2. You are eliminated as soon as you get 2 loses.
As soon as you win you go to the report station with the match slip. You will be paired up against another player, given the match slip and the table number. The standings won't matter because as soon as you lost twice you are out of the tournament already.
No draws for first day what so ever. If you fail to win X matches when the given time comes, you don't make day 2.
Of course, this will only work for first day.
Wanna hear what I think about restaurants?
Check out my http://damancy.blogspot.com/
Trust me! IM FAT!!!!
I personally don't think it's worth "solving" the line items that you listed that are not actually problems.
"no draws" is impossible. Earthquake for 50, what happens? Also, if it were possible to institute "no draws", then what the point of changing from swiss?
As for "you finished, move to next available opponent" Years (1998? 97?) ago, I joined a large tournament with something similar to this.
What happened was several players just played 1 game, then turned in their slip, with the agreement that who won the first game also won the second.
By the time I finished 3 games and realized this, some players were already on their 8th or 9th. The time was called, I lost 1 game, won 6, but there were players who won 15+ matches. Landed 30th place or something.
Also, "lose 2 games, you go home" doesn't address the problem mentioned by previous posters.
Finally, if you institute a "win X to go to day 2", people who already won X will simply stop playing... which _will_ end up with players automatically locked out because they no longer have any opponents left, even if there is time left or they have no losses.
"Sometimes, the situation is outracing a threat, sometimes it's ignoring it, and sometimes it involves sideboarding in 4x Hope//Pray." --Doug Linn
Taking a sporting example, DE could work in Australian Rules football (where scores are usually around 90 a side making ties uncommon) but not in soccer (where a good quarter of matches between teams of comparable calibre end as draws).
Actual draws in Magic are rare but for time reasons they are common, and stalling is a real issue as a result.
The only solution I can think of is to have a system where you are playing DE, but in the case of a timed-out match you have a much more ruthless way of determining the winner - either via an in-game sudden death scenario (as the tournament rules allow for in the rare case of a single elim scenario where time runs out) or via an out of game system (at the end of the time period, if the game is undecided, it's decided by a coin flip).
The 'being bored due to early elimination' issue isn't a problem as DE eliminates people in big groups and in predictable numbers, so side events would be easy to organise. If you start with 512 players, DE will eliminate 128 after round 2 and a further 128 after round 3 and a further 96 after round 4 - all in big groups.
I assume there would be a match limit number. As in you get 8 rounds total to work with and need at least 6 wins to make day 2.
The no draws in this scenario is somewhat easier as the magic tournament structure already does this. Each match is to the best of two which in a match without a time limit like these would would just mean you go to game 3,4,5, 6 as necessary.