|
On February 18 2015 03:54 opisska wrote:Show nested quote +On February 18 2015 03:36 Xoronius wrote:On February 18 2015 03:28 opisska wrote:On February 18 2015 02:12 The_Red_Viper wrote:On February 18 2015 01:40 opisska wrote:On February 18 2015 01:26 The_Red_Viper wrote:On February 17 2015 22:39 opisska wrote:
How can you even think to say that RR is fairer? Somebody even brought up the problem, but it was quickly dismissed, because it is inconvenient. The situation when somebody's advancement is decided by a match of two other people, one of which has nothing to gain/lose is inevitable in RR and it is the polar opposite of "fair". You can toss around "professionality" as much as you want, it won't make it fair. What case are you exactly talking about? I am talking exactly about the case I have so explicitly described. Not a specific instance of it happening, but about the fact that this can happen and cannot be avoided. so you are talking about mapwins as decider? (kinda what TB was angry about in a dreamhack some time ago) Ok, let me give you a specific example, right? A>B C>D A>C B<D A<D A 2 B 0 C 1 D 2 now we play B vs. C - if B wins, A and D advance while if C wins, then A>C>D>A is a 3-way tie at 2 points. But there is nothing for B in the last game (except for pride) and that gives C a huge and unfair advantage. He is not out of the woods yet (in more than 4-player groups you can find examples when it's a straight up advancing, not a tie, but I couldn't find one for 4 players), but definitely better off than if he lost ... If GSL groups, every match is the players plying for themselves, a player never plays in a situation where his faith (good or bad) is already decided. In RR there are always 2 matches played at the same time. Thus A vs D and B vs C are played at the same time and B doesn't know, that D will beat A. Thus the standings would be: A 2 B 0 C 1 D 1 And B would still have a chance to win a tie-breaker, thus he would play for something. In the hypothetical case, that the result of A vs D would be known (which is, as said before, usually not the case), the problem is still easily fixable: Make the money distribution 4,5k/3,5k instead of 4k/4k and suddenly instead of playing for nothing, B plays for a chance to maybe overtake C and thus for 1k. 1. in a broadcasted league, you will probably want to play the matches one after another because of the viewers. I think there is some space to help things here, because you can choose the order, but the problem is that that also influences the matches, so is not so "fair". 2. what? if tha A-D results is known, B does not advance no matter what, so the prize money distribution is not helping at all. You can make per-match prizes to give incentive, but that can turn out to be expensive and also it is very different to play for 100$ and for a shot on the GSL trophy.
1. A&B-stream. It has been done in the past a lot
2. B wouldn't have a chance to advance in said scenario, but he would have a chance to get 3rd place. If the difference between 17-24/25-32 is 1k$, that might be 1k$ for winning a single bo3. That is about 500-1000$ per hour. I think most people would work for that amout of money
3. _TMT_ problem with double-elim is still up and hasn't been answered.
|
1. it has been done at weekend tourneys. In GSL I fucking want to see all of the games! Mind you that all of this business is kind of done for the spectators in the first place, without them, there are no prizes, no tourneys ...
2. fair point, but you need to see the huge difference. Also opens a barn door to fixing (sorry to pull that up in these days, but it's the reality)
3. I am sorry but this is really subjective. All of this "mental exhaustion" and "show cards" is highly speculative. I believe that every player should be able to do for himself in the current match as well as he can (that holds IRL as well for that matter) and the rest are really just excuses. The problem in my example is that A and D don't really have much chances to do anything for themselves in the last match where they are not even playing.
The important thing to note is that neither method is without issues and it's a lot about which kind of issues does one see as more important. But shouting that RR is "fair" is really misguided. The only absolutely fair thing would be a repeated RR with random ordering of matches going on for an infinite number of rounds to smear any assymetries due to match order (and even that would be terribly OP against mortal beings ...  )
|
On February 18 2015 04:12 opisska wrote:1. it has been done at weekend tourneys. In GSL I fucking want to see all of the games! Mind you that all of this business is kind of done for the spectators in the first place, without them, there are no prizes, no tourneys ... 2. fair point, but you need to see the huge difference. Also opens a barn door to fixing (sorry to pull that up in these days, but it's the reality) 3. I am sorry but this is really subjective. All of this "mental exhaustion" and "show cards" is highly speculative. I believe that every player should be able to do for himself in the current match as well as he can (that holds IRL as well for that matter) and the rest are really just excuses. The problem in my example is that A and D don't really have much chances to do anything for themselves in the last match where they are not even playing. The important thing to note is that neither method is without issues and it's a lot about which kind of issues does one see as more important. But shouting that RR is "fair" is really misguided. The only absolutely fair thing would be a repeated RR with random ordering of matches going on for an infinite number of rounds to smear any assymetries due to match order (and even that would be terribly OP against mortal beings ... )
3. Sorry but i don't think that this answers any of my points.
The problem is that:
A & B both play 2 matches before their final clash.
One match they play each other.
One match A plays against the best, and B plays against the worst.
So coming to the final match, the advantage is CLEAR.
And no, one does not come to a match "do as well as he can". It all comes down to builds and preparation. Especially with this GSL-like format. And if you know winning does not matter (losing even benefits), why not hide your builds in the first match and let your opponent shows all his?
Double benefit here.
The Rematch curse has existed for so so long (so annoying too) and organizers still think it's ... a curse????
P/S: Your arguments are all about random factor. Grouping is random. Ordering is random. So it's fair after all.
One can not roll a bad dice and says it's not fair
|
Not going to argue over 3, since that _TMT_ is doing that, maybe I'll say something at a later point.
On February 18 2015 04:12 opisska wrote:
1. it has been done at weekend tourneys. In GSL I fucking want to see all of the games!
1. GSL is GSL and GSL might be on TV again at some point and they have to furfill a schedule there, so I can accept GSL sticking to their own format, because of scheduling reasons. But the double-elim stuff is everywhere at the moment, even if those scheduling reasons are not existent.
On February 18 2015 04:12 opisska wrote:
Mind you that all of this business is kind of done for the spectators in the first place, without them, there are no prizes, no tourneys ...
In that case, we have different expectations on what a systems primary point is. I care about optimal fariness to the players, when judging a system and about nothing else. Spectators are totally negligible from my PoV. (I see, where my view collides with the majority here though, I also still don't see a problem in the Nani/Polt-incident). Take that one MLG thing, where everyone picked Puck for example: Nice storyline, fun to watch for the viewer, but undoubtly unfair and thus IMHO a bad system, destpite it's entertainment value.
On February 18 2015 04:12 opisska wrote:
2. fair point, but you need to see the huge difference. Also opens a barn door to fixing (sorry to pull that up in these days, but it's the reality)
I'm not seeing, how it would do that. Please elaborate, because if there is a huge point here, that would be intresting.
|
1. I concede that tourneys who schedule two streams anyway could in principle use RR. I couldn't create a 4-man scenario when the problem would occur without knowing the result of the other match. However you should lose the tunnel vision about "players" - optimal for them is in no doubt to get some money from the match and that's directly dependent on the viewer popularity of the tourney. That's just the reality of life I am afraid.
Also I hope that we could at least agree that tourneys should not use >4player RR, right? Because there it gets really, really messy fast.
2. I think this is pretty straightforward, albeit nasty - a player who has nothing to gain could lose on purpose and the advancing player could cut a part of his prize money for him. I am not really trying to accuse anyone of such behaviour, but the opportunity makes the thief. This aside, I still see a big difference of motivation in playing for a shot on the trophy and for a few bucks (in the case your financial incentive solution is used).
3. ad _TNT_ - I am sorry man, but all of this "point" is just a story you made. You purposedly overvalue one of many aspects of the competition to create tension in your scenario. First of all in high-level SC2 there are hardly situations where players of definedly different levels like that.
Also if I wanted to play along with you, I would say that A can just throw (or even forfeit, I think this is allowed in WCS according to the Book of Aeromi) the games against H which he would lose anyway and relax, gaining an advantage over B for the rematch (because B has to at least try to win over L). But please don't cite me on this, it would mean that I agree with your theory that you can put such "skill labels" on players which I do not.
|
On February 18 2015 19:54 opisska wrote:
1. I concede that tourneys who schedule two streams anyway could in principle use RR. I couldn't create a 4-man scenario when the problem would occur without knowing the result of the other match. However you should lose the tunnel vision about "players" - optimal for them is in no doubt to get some money from the match and that's directly dependent on the viewer popularity of the tourney. That's just the reality of life I am afraid.
Also I hope that we could at least agree that tourneys should not use >4player RR, right? Because there it gets really, really messy fast.
2. I think this is pretty straightforward, albeit nasty - a player who has nothing to gain could lose on purpose and the advancing player could cut a part of his prize money for him. I am not really trying to accuse anyone of such behaviour, but the opportunity makes the thief. This aside, I still see a big difference of motivation in playing for a shot on the trophy and for a few bucks (in the case your financial incentive solution is used).
3. ad _TNT_ - I am sorry man, but all of this "point" is just a story you made. You purposedly overvalue one of many aspects of the competition to create tension in your scenario. First of all in high-level SC2 there are hardly situations where players of definedly different levels like that.
Also if I wanted to play along with you, I would say that A can just throw (or even forfeit, I think this is allowed in WCS according to the Book of Aeromi) the games against H which he would lose anyway and relax, gaining an advantage over B for the rematch (because B has to at least try to win over L). But please don't cite me on this, it would mean that I agree with your theory that you can put such "skill labels" on players which I do not.
Yes that story is so made-up, so overvalued, so rare that almost EVERY tournaments have it now.
Of course, you can ignore it, call it a curse maybe
|
I'm just going to drop it : GSL group format with Extended Series
|
|
|
|
|
|
EPT
