ETF2L – Latest activity in “Why Does GC Rule Still Exist?”

ETF2L – Latest activity in “Why Does GC Rule Still Exist?” https://staging.etf2l.org/forum/league/topic-31027/ Reply by Admirable https://staging.etf2l.org/forum/league/topic-31027/page-2/?recent=526147#post=526147 generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526147 Mon, 02 Feb 2015 16:51:31 +0100 Reply by Fuxx https://staging.etf2l.org/forum/league/topic-31027/page-2/?recent=526147#post=526146 Quoted from Starkie

[...] no its not

ye it is]]> generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526146 Mon, 02 Feb 2015 16:40:02 +0100 Reply by Permzilla https://staging.etf2l.org/forum/league/topic-31027/page-2/?recent=526147#post=526143 Quoted from Max

[...] Wait that means a strong team might have 0 points while a weak team might have 2 points after 2 weeks Then what happens? The strong team gets matched up with other weak 0 point teams and the "no weak vs strong" thing you are trying to get away from is actually coming back. However if that is the risk you wanna take then ofc feel free to. However how do you think you will exactly determine who is weak and who is strong?

http://en.wikipedia.org/wiki/Swiss-system_tournament#Accelerated_pairings We are seeding them ourselves, based on history/screenshots etc. Just like we would seed last season.]]> generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526143 Mon, 02 Feb 2015 16:01:42 +0100 Reply by HedoKingoglu https://staging.etf2l.org/forum/league/topic-31027/page-2/?recent=526147#post=526141 Quoted from Permzilla

[...] We are using the accelerated pairing system which matches strong with strong and weak with weak in the first two weeks.

Wait that means a strong team might have 0 points while a weak team might have 2 points after 2 weeks Then what happens? The strong team gets matched up with other weak 0 point teams and the "no weak vs strong" thing you are trying to get away from is actually coming back. However if that is the risk you wanna take then ofc feel free to. However how do you think you will exactly determine who is weak and who is strong?]]> generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526141 Mon, 02 Feb 2015 15:56:17 +0100 Reply by enzyme https://staging.etf2l.org/forum/league/topic-31027/page-2/?recent=526147#post=526134 Quoted from Herr_P

[...] 10/10 excellent rebuttal

no its not]]> generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526134 Mon, 02 Feb 2015 14:57:55 +0100 Reply by Herr_Pyromatic https://staging.etf2l.org/forum/league/topic-31027/page-2/?recent=526147#post=526125 Quoted from Fuxx

no ur not

10/10 excellent rebuttal]]> generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526125 Mon, 02 Feb 2015 09:50:45 +0100 Reply by Fuxx https://staging.etf2l.org/forum/league/topic-31027/page-2/?recent=526147#post=526122 generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526122 Mon, 02 Feb 2015 05:37:00 +0100 Reply by Permzilla https://staging.etf2l.org/forum/league/topic-31027/page-2/?recent=526147#post=526121 Quoted from Max

[...] As far as I know swiss system matches weak teams against strong teams in the first few matches. So I dont really think trying to get fair matchups is what you are supposed to do as it is hard to do and it ruins the nature of swiss tournament system.

We are using the accelerated pairing system which matches strong with strong and weak with weak in the first two weeks.]]> generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526121 Mon, 02 Feb 2015 03:48:58 +0100 Reply by fraac https://staging.etf2l.org/forum/league/topic-31027/page-2/?recent=526147#post=526120 generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526120 Mon, 02 Feb 2015 03:48:43 +0100 Reply by HedoKingoglu https://staging.etf2l.org/forum/league/topic-31027/page-2/?recent=526147#post=526119 Quoted from ducky

[...] Determining who's higher on the table at the end of the season is not an issue,we can do that with the current point system we have using tiebrakers/head-to-head/round difference/sonneborn-berger-score The matter at hand is to generate as accurate and balanced matchups as possible for the teams as early as the very start of the next week. To do that effectively we need a clear winner at the end of the previous week. Tied matches resulting in equal points for both teams won't cut it. Someone HAS to win,whether it's a 3-0 map win or a 2-1 GC win.

As far as I know swiss system matches weak teams against strong teams in the first few matches. So I dont really think trying to get fair matchups is what you are supposed to do as it is hard to do and it ruins the nature of swiss tournament system. That is what the average system I suggested comes in. In chess they matchup players according to their score and their elo's. Since we dont have elo's our only chance of getting a natural matchup is to have some sort of an average system going on. Either the one I suggested or round differences. Or we could have none and only look at points when matching up. I can also see where you're coming from with "someone has to win" but that means we have to remove the gc point difference]]> generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526119 Mon, 02 Feb 2015 03:25:07 +0100 Reply by fraac https://staging.etf2l.org/forum/league/topic-31027/page-1/?recent=526147#post=526117 generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526117 Mon, 02 Feb 2015 02:40:45 +0100 Reply by dnscodcs https://staging.etf2l.org/forum/league/topic-31027/page-1/?recent=526147#post=526116 Quoted from fraac

But if Team A and Team B draw, does it matter which of them is scheduled to play Team C? Surely a draw says they're interchangeable.

I guess this is assuming that we make use of a point system that is able to divide points for draws equally between the teams,like the 0,5 point system that porter suggested? Let's say you start from week 1: In a balanced environment teams who tied their week 1 match will only face equally skilled opponents for the next week if they're placed against teams who've also tied their week 1 match. So if a draw occured in a week and we want to make balanced fixtures for the next week we'd need at least 4 teams drawing their matches in that previous week. Same goes for any week after that. If it's more than 4 than it'd have to be in numbers dividable by 4 e.g. 4,8,12,16 etc. so that there's enough tied teams to place against each other for balanced matchups. If that's not the case and since there's only so many drawing teams to place against each other you'd have to create fixtures that are unfair for one of the teams (e.g. a team who tied their week 1 match would end up getting placed against a team who won/lost their week 1 match in their week 2 match,resulting in an uneven seed either way) To illustrate: Team A draws Week 1 with Team B Team C wins Week 1 against Team D Team E wins Week 1 against Team F Team G wins Week 1 against Team H So the Week 1 ending point score is: Team A: 0,5 Team B 0,5 Team C 1 Team D 0 Team E 1 Team F 0 Team G 1 Team H 0. Placing the bottom 4 teams is easy and results in balanced matchups,since you only have to place winners vs winners and losers vs losers (Team E vs Team G , Team F vs Team H) However the top 4 teams will have rather random and unbalanced matchups no matter how you look at it. Team A or Team B will face a stronger or weaker opponent for their Week 2 match (Team C or Team D). And this isn't just an occurence for the Week 1 or Week 2 matches,if you do the math you will figure out that for this kind of system to be balanced you're pretty much counting on a certain number of draws every week onwards just to even out one or two tied match/es that happened in a previous week. TL;DR: Having a clear winner makes all of this easier by eliminating the need for hoping on certain match outcomes just to create balanced fixtures.]]> generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526116 Mon, 02 Feb 2015 02:29:05 +0100 Reply by fraac https://staging.etf2l.org/forum/league/topic-31027/page-1/?recent=526147#post=526114 generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526114 Mon, 02 Feb 2015 00:33:35 +0100 Reply by dnscodcs https://staging.etf2l.org/forum/league/topic-31027/page-1/?recent=526147#post=526110 Quoted from Max

... Maybe this as a point system can work in future seasons?

Determining who's higher on the table at the end of the season is not an issue,we can do that with the current point system we have using tiebrakers/head-to-head/round difference/sonneborn-berger-score The matter at hand is to generate as accurate and balanced matchups as possible for the teams as early as the very start of the next week. To do that effectively we need a clear winner at the end of the previous week. Tied matches resulting in equal points for both teams won't cut it. Someone HAS to win,whether it's a 3-0 map win or a 2-1 GC win.]]> generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526110 Mon, 02 Feb 2015 00:00:52 +0100 Reply by Permzilla https://staging.etf2l.org/forum/league/topic-31027/page-1/?recent=526147#post=526107 Quoted from Max

1 point per map win 1/2 point in a tie (no GC) Average is determined by how many points in total the teams you've played against has. example: Team A and Team B has the same points Team A has played against Team X, Team Y and Team Z Team X has 5 Team Y has 5,5 Team Z has 6,5 at the end of the season Team B has played against Team C, Team D and Team E Team C has 4 Team D has 4,5 Team E has 2,5 Since Team A has played against teams that had higher points at the end of the season, they have 17 average points Team B has 13 average points Team A is higher on the table because of higher average. Maybe this as a point system can work in future seasons?

What you've described here is a very similar version of the Sonneborn-Berger score, which is the 2nd tiebreaking method we're using this season.]]> generator=rsdiscuss&baseurl=https://staging.etf2l.org&feed=forum&forum=league&topic=31027&post=526107 Sun, 01 Feb 2015 23:46:55 +0100