|
What most people dont know/understand is that extended series actually makes it less likely that the best player in a field actually wins the tournament compared to in a normal double elemination bracket format.
Im not pulling this out of my ass and saying this based on some asumptions im personally making. This statement is based on mathematics and advanced theory that I am not going to post as a comment here.
Ill tell you the conclusion and basics of it is this:
Consider every starting-slot in a double elemination bracket occupied by a precentage instead of a player.
The combined precentages of all players will be 100% so the starting precentage for every single bracketslot will be 100/<numberofslots> since we are not considering any games outside of our own bracket, nor having seeds etc this is how it will be.
Now the precentage represents a players chance of beeing the best player within the field. (Assuming this is what we are trying to determine with our bracket, and the more accurate we determine this the better our bracket system is.).
Now imaging these precentages playing eachother and advancing in the bracket.
The numbers will drop and rise as they progress into the bracket depending on their results.
How do we determine how effective our bracket is at finding the best player within a field of players?
By the precentage that the Winner of the bracket ends up with. If we had a system that would give every eliminated player a precentage of 0% and the Winner of the bracket 100% then this would be the perfect bracketsystem.
This does not exists!
However, I can tell you that the double elemination bracket with extended-series system would produce a significantly lower precentage in the Winner-slot then a normal double elimination bracket would.
Wich to me indicates that extended series doesn't make sense for a bracket with the purpose to determine the best player within a field as accurate as possible to use extended series.
Lets look at some logistics using the same precentag-bracketmap as you imagined previously.
Now I will use a slot deep in the lower bracket for a demonstration and I will be using the "lower-backet semi-finals" (in a double elimination bracket there is only 1 semi-final in the lower bracket, and the Winner of this game will play the loser of the upper-final in the lower-final.).
The precentage assigned to the player in this slot will ALWAYS be ROUGHLY the same, the path taken to get there will vary extremly slightly and the ammount is so small that it's irrelevant.
Now lets think about what that means!
It means your precentage will always be the same if you reach this point of the tournament and it doesnt matter if you lost round 4 upper and won a few games in lower, or you lost round 6 to get placed in the lower-finals. So why would you sometimes give one player a lead over the other player in lower-semis based on if they have played eachother earlier in the bracket or not? Remember you are playing precentages and the player behind the precentages doesn't actually matter.
In this example one player lost in round 4 upper match, while the other won his round 4 upper match, it shouldnt matter to WHO you lost in round 4 upper (since they are all the same precentage.) when you are in lower-semis.
Remember that the purpose of the bracket IS to determine the best player within a field as accurate as possible.
Remember that the best player in a field is the player thats favored over the most other players in the field, this isn't often every player in the field.
Remember that the purpose of the bracket ISN'T to determine as accurate as possible who is favored in any given player collision.
We can clearly see that the purpose of the extended series rule clearly deviates from the general purpose of a bracket and also produces a "less legit" champion.
Now, If people consider exnteded series more entertaining then using a normal more fair rule-system then I can see this beeing used but I simply can not imagine that the audience like exntended series better. I think the implementation of extended series comes from an uneducated asumption that it would produce a more fair and legit bracket while actually it is the drastic opposite!
|
I don't like extended series because if a player beats another player then the player that won the first game shouldn't have any trouble winning again
|
i dont like the idea that if player A beats player B and then later player A has a significantly higher chance to progress vs. player B than vs player C.
or imagine B beats C and then loses to A. and before that C beat D. what kind of fucked up series are ahead?
so C meets D in lower bracket and is more likely to advance because he has this possibly huge point advantage.
after that C plays vs B again, this time from far behind.
so then player B comes out of the loser's bracket and "fought" his way back (yeay he won one game) and then he doesn't have to pay the price that others did when facing A again in the finals? either way it's bullshit.
you think the player should have an advantage because he won vs another player before in the tournament?
he already has one, the pressure is on the opponent.
|
I don't mind extended series too much tbh, but I also can see why people don't.
I'm just curious why they decided to put it back in...
|
its not a big deal imo. it rarely happens throughout the whole tournament and I can definitely see its purpose
|
I hesitate to resurrect this because it's such a polarizing topic, but I feel like one of the things extended series wants to do is reward someone for winning 2-0 instead of 2-1 (and to prevent a net 3-2 record against a player leading to elimination). I still think that could be solved with a 1-0 start in BO5 for a previous 2-0 result and an even BO3 for a previous 2-1. I think this leaves a lot more room for either player to win - needing to go 3-1 (or 4-1) is just really unlikely.
If the goal is to prevent someone from winning first and then losing to the same player feeling unlucky for losing at the wrong time, then MLG really needs to have losers go to different brackets to reduce the chance of meeting again.
To whoever suggested a swiss system, I think it'd be super cool to try a tournament like that but you'd need to play at least 10 2 game rounds to get any kind of separation at the top - you might even need 15 before it stops being just as chaotic (if not more) as MLG style double elim.
I just thought I'd throw my bad ideas out there now since we're finally in range of extended series tomorrow morning.
|
On June 25 2013 09:08 ValhallaDude wrote:
I don't see why so many people have this purely emotional aversion to the extended series.
If a player beats another player earlier on in the tournament and goes on to win all matches and he faces that player again in the finals, why would it not be an extended series?
Why does the player who wins get a double jeopardy and the player who loses get a second chance?
We don't try people for the same crime twice. This should be no different.
I don't think you understand what double jeopardy is... The finals of MLG will be between the winners of two different brackets, so it'll obviously make no sense to have extended series cause there will be a 0% chance that they could have met earlier. The aversion people have to extended series in the finals is that is puts the person in losers in an even more precarious position. Aside from that, it eliminates the possibility of the always exciting "bracket reset". In a Grand Finals, the person in losers (regardless of whether they ever met the person in Winners before the finals) is required to win two series instead of one. Because it's double elimination, the person in Winners only has to win one series out of two, giving them a massive advantage but still allowing the person in Losers to have a chance by outplaying his opponent. This is more fair and, more importantly, more exciting for viewers than someone coming in to a series down 2-0 from the start. It shouldn't matter who they lost to in Winners or what score they did, it just matters that they lost. That's why they are in Losers.
|
So, apparently, the extended series rule is in effect during the MLG spring championships? I don't see any mention of it on the Liquipedia tournament page.
Round 8 spoiler
+ Show Spoiler +So, let's say Jaedong beats Naniwa ,his match against Dear will continue from 0-2?
|
yes, polt is 2-0 against naniwa already
|
I just liked MLG back when it would be a RO32 in the Championship series. Im not a huge fan of the RO128 with a bunch of random American players. I was thinking about going to Anaheim a long time ago, but i decided not to go since of the weird format. :/
|
annoying.. takes the hype off, but its fair
|
The naniwacase yesterday shows the problem with extended series He did lose to Polt and the result was that he had to win three matches to get a place in the semi-finals, with Polt only having to win one more match. Despite the three strong wins he was still punished in the semi for his first loss which strongly limited his chances and made it less exciting because of the desperate situation of having to win 4-1. In other words a double punishment for a loss that is unreasonable.
This is cold facts, nothing to debate further really. Extended have proved unfair.
|
On June 29 2013 05:54 Tanngrisnir wrote:
...I think the implementation of extended series comes from an uneducated asumption that it would produce a more fair and legit bracket while actually it is the drastic opposite!
well said!
also I think MLG does it just to be different than other tournaments
|
Yeah, it crossed my mind that they do it just for publicity. The loss to the integrity of the tournament is not so large, and here we are talking about MLG. And we will be talking about it long after the tournament.
I dont mind, extended series in final of double elimination bracket (they have two doble elim brackets as part of in MLG final bracket right?). What worries me is extnded seires before fnal, when one play has advantage over the other, while they both are in losers bracket. Is it possible in current form of MLG rules? If so thats bad.
|
I really don't see the big deal about it. The only thing they should have done was make the finals a Bo7 instead of a Bo5 since Hyun didn't lose a series at the tournament yet. Extended series is fine.
|
it's totally fair, people will always qq about it when their favorite player loses in an extended series (even though it is better proof that the better player won than a bo3).
deal with it
|
The advantage of not getting knocked down to the loser's bracket is that you don't have to play the matches in the loser's bracket. What's the point of giving you even more of an advantage? You're going up to the guy who just fought tooth and nail through the loser's bracket and then they have a disadvantage going into a series? On a match by match level, every matchshould be equal. No questions asked. Let the better player win. The extended series, if anything, just makes it so the better player doesn't always win because if somebody had an easy road to the finals and didn't get knocked down to the loser's bracket they could win against someone who is better but had a much harder road and did get nkocked down.
|
Never been a fan of extended series. I prefer each series being its own independent event. No need to link them together because you happen to be playing against the same guy again.
|
Extended series so good idea. What about making extended series between two or three following turnaments...
|
If the players meet up again it should just be a normal game. I don't think it's very fair to give a player such a huge advantage, and to be honest, it's boring as well. Other tournaments don't do that either.
Imagine this in the GSL finals "oh player X has a lead because he beat player Y in the RO32. Player Y now has to win 4 games in a row, while player X only need to win twice". Everyone would flip their shit.
|
|
|
|
|
|
EPT
