|
On July 24 2013 09:30 renlynn wrote:Show nested quote +On July 23 2013 07:19 StarStruck wrote:On July 23 2013 07:07 kingNothing42 wrote:On July 23 2013 06:12 Day[9] wrote:
To clarify, we are not saying that worse players have a statistical advantage overall. Rather, in a double elimination bracket, most of the extended series will be a worse player beginning with a lead over a better player.
That's a pretty good point. However, I don't think we should start Series#2 ignoring GOOD's mistakes. GOOD flubbed Series#1. Why should we give him more of a chance to overthrow BAD if BAD legitimately won Series#1? I mean, it's not like BAD cheated to beat GOOD. The concept that BAD drops out of the bracket a lot more commonly than GOOD is true. But by the same point as above, why should BAD get an advantage against GOOD if they happen to meet again (albeit less likely than the converse) just because he stuck around? The conclusion is that Bo3 makes for a more volatile tournament. That's may be good for spectators. But saying that we should give the GOOD player an advantage by giving him/her a second Bo3 is not convincing to me. His/her chances are better in a Bo7. It's not like players never come back from the disadvantage (cough Soulkey cough). That usually happens if they're really, really good players with a strong playbook. You know who also came back in extended series? Leenock. He was on fire that time though. If someone gets red hot at a LAN good luck taking them out. if you're referring to him winning the finals from the lower bracket, that wasn't coming back in extended series, because he had no prior history there. that was him benefiting from the one time in MLG where extended series favors the lower bracket. normal double elimination would reset the score after the first series, extended series in the finals does not. I guess this could be an argument in favor of extended series since it makes things more even in the finals. sometimes. unless you drop 2 games at the start. but only at the start, dropping 2 games in the middle is still okay.
Are we talking about the same thing? DRG beat Leenock 2-0 in Winners Round 2 of MLG Providence in 2011. They would later meet in the losers Final where Leenock beat him 4-3. -_-
|
oh, sorry. forgot about that.
|
lets all be honest here, as a "viewer". Not taking to account all the factors say: Players condition, MLG's Budget and all that jazz. We can all agree that the more games we can watch the happier we are. Do we all agree with this?
So with that even if more games would result to some shitty games of cheese,giving changes to prove ones worth or just giving a statement by providing quality games from a player, we as viewers are still all in favor in more games.
The advantages (as far as I can only remember and list down:
1. More games, more replays more material more fun? 2. Will prove a player is indeed worthy of the bout (self worth and bragging rights, makes the person really prove he is better than the other player) 3 ? Will give a chance to the early lossing player (well only advantage was for him) - he might be just nervous and needs more time to cope with tounament fright? Can still prove himself? Can recover from cheesy stuff? Can learn and become better in the next encounters?
The disadvantages:
1. Could result to some games to cheese ( but still can happen to shorter bouts) 2. Will tire players ( we can do a week long tourney which results to number 3) 3. Budget ( higher tickets? More sponsors? less price money? It is MLG's problem?)
Therefore I can say, the more bouts, lossers bracket/ extended series etc = better event. Viewers wins!
|
On July 24 2013 23:36 woreyour wrote: lets all be honest here, as a "viewer". Not taking to account all the factors say: Players condition, MLG's Budget and all that jazz. We can all agree that the more games we can watch the happier we are. Do we all agree with this?
...
Therefore I can say, the more bouts, lossers bracket/ extended series etc = better event. Viewers wins!
More games != more fun Better games = more fun and better viewer experience
Look at my example, last post page 4. If the BAD player ends up winning the extended series (knocking out the GOOD player) He will very likely soon meet an opponent of a race he is bad against, giving really really bad games forward.
(Assuming games between GOOD players are more fun to watch)
Best example of getting far through one good matchup is the GSL season were Inca reached the finals vs Nestea (Through only playing PvP's iirc). Hopefully we can all agree, that was one of the worst finals ever played,
|
I would say more games are a disadvantage if it doesn't help us determine the objectively best player (if there is such a thing). Making the format more entertaining isn't a bad thing, but if we're going to put more games played into a tournament, is it really a good idea to stuff these extra games into the lower bracket? The lower braket is already a bottleneck when it comes to scheduling. If we wanted more games, why not make the upper bracket games BO5 instead of BO3? That would give us more games with people who are actually winning while making it less likely better players fall into the lower bracket on fluke results.
|
I don't understand the big fuss about all this anyways. If MLG believes that in order to win their tournament, they must win the majority of games against everybody they play, that's absolutely their perogative. It's OK for two tournaments to have two different philosophies on this matter. If Dreamhack or IEM or ESL or NASL has a different philosophy, that doesn't mean MLG is wrong, just different. It's their tournament, they're spending the money to run it and stream it, they get to say how a player advances.
It's just an excuse for people to hate on MLG more. We all know how much people love that.
|
On July 23 2013 06:12 Day[9] wrote: The result in the original post might surprise people, but it makes sense when you consider the implications of player skill in a broader bracket. Consider the following:
Suppose we have two players that I shall name GOOD and BAD. Further suppose that I am omnipotent and can determine with full accuracy that GOOD beats BAD exactly 70% of the time. We place these two players into a double elimination tournament with extended series, each round being best of 3.
Scenario 1: GOOD beats BAD in the winner's bracket (expected result). BAD drops to lowers. Because BAD is unfavored against GOOD, it is extremely likely that BAD is also unfavored against other players as well. Because GOOD is favored against BAD, it is extremely likely that GOOD is also favored against other players as well. As a result, BAD has a high probability to be knocked out of the tournament in lowers long before he ever meets GOOD again. The simple conclusion is thus: is GOOD beats BAD in the winner's bracket, there is a low probability that an extended series will even happen. For the sake of argument, lets say there is a 5% chance that GOOD meets BAD again. When this does occur, GOOD has a very high chance to beat BAD as a result of the extended series setup (GOOD begins with a lead).
Scenario 2: BAD beats GOOD in the winner's bracket (unexpected result / the "fluke"). GOOD drops to lowers. As we said before, because BAD is unfavored against GOOD, it is extremely likely that BAD is also unfavored against other players in the tournament. Consequently, BAD has a high probability of falling to lowers sooner rather than later. Similarly, since GOOD is favored against BAD, GOOD has a high probability of advancing through lowers. Therefore, there is a much higher probability that GOOD will meet BAD an extended series will happen. Lets suppose there is a 20% that GOOD meets BAD again. When this does occur, BAD has quite an edge due to the extended series setup (BAD begins with a lead). So, although GOOD is favored in an individual match against BAD, BAD still has a higher probability of winning in an extended series.
Based upon these (somewhat winged) numbers, we see that, when an extended series DOES occur, MUCH more often it is a bad player starting with a lead against a good player. So, "worse players" will win more often in an extended series double elimination bracket.
To clarify, we are not saying that worse players have a statistical advantage overall. Rather, in a double elimination bracket, most of the extended series will be a worse player beginning with a lead over a better player.
This totally changed my mind on this subject and IMHO should be added to the OP. I was (and still am) a huge proponent of the extended series. Mostly because I think it is more fair. And there is just the added drama of the Bo7 when the better player is behind (one of my favorites to bring up is MC vs Idra). It just seems unfair to me that someone GOOD could lose 2-0, then come back and win 2-1 in the loser bracket and advance over BAD with a total map score of 2-3. Reading your analysis has convinced me that the extended series is putting GOOD at a disadvantage to BAD, which probably should be enough to get the whole system scrapped.
|
On July 23 2013 22:53 KissMeRed wrote:Show nested quote +On July 23 2013 18:23 ZeroPageX wrote:On July 23 2013 09:58 KissMeRed wrote:I've written a highly stylized story to convince everyone that Extended Series is the best rule ever implemented in tournament play. We're at the 2014 MLG Winter Championship in Detroit, Michigan where 128 SC2 players from around the world are competing in the open bracket. At the request of a vocal internet minority, MLG has decided to conduct the tournament without the standard extended series rules. The player pool in uncharacteristically weak featuring only two Koreans. However, fans are still excited since they will get to witness some epic foreigner beatdowns coming from the hands of KT Rolster's Flash and Samsung Khan's Shine! Fast forwarding through the E-sports massacre, we are in Winner's Round 6 featuring none other than our two Korean heroes, Flash and Shine! Now Shine knows he can't possibly beat Flash in a standard game (0% win rate on ladder T.T), but he's been saving two of his most devious Zerg cheeses specifically for MLG Detroit! These cheeses are guaranteed wins for Shine! Since there is no extended series, and because he's smarter than Flash, Shine realizes the only way he can win this tournament is to purposely lose in Winner's Round 6 and save his cheeses for the impending Semifinals rematch. Once Flash sees the build orders, they will become useless. Shine executes two of his drones in Game 1 to go for a 4 pool and attempts a mass overlord 'blinding' strategy in Game 2, both of which end in losses. Flash take the series 2-0. Shine makes quick work of a non-Korean chump in Loser's Round 10 and gets ready for the semifinals rematch. Remember, Shine has two cheeses that are 100%, guaranteed map wins against Flash. Shine goes for his first cheese on Newkirk Redevelopment Precinct (yes, it's still in the map pool). MLG cuts the video feed halfway through the match, it's too much for the audience to handle. Shine wins Game 1. Game 2, Flash suffers the most horrifying loss of his career on Red (Sick) City and instead of typing the customary 'gg', he plunges his ruler into the face of the monitor in a fit of rage. Shine wins the series 2-0 and advances to the Grand Final! Flash is eliminated! The Grand Finals are about to begin, but there is one more twist. Shine and his competitor, ID: STXUncleDrew, shake hands. UncleDrew begins peeling back professional grade makeup to reveal that Shine's opponent is actually INnoVation in disguise! Men, women, and children all weep uncontrollably since they don't get to see a Flash vs Innovation finals. Then they log-on to Teamliquid.net to comment on a thread about why MLG should have upheld their extended series policy. The end. (Disclaimer: This is not a true story or vision of a real future. No disrespect to any players, maps, or tournaments referenced above.) How did Flash wind up in the losers bracket? He never goes to the losers bracket. I followed the most recent MLG format: http://wiki.teamliquid.net/starcraft2/2013_MLG_Spring_ChampionshipWhen you lose in the semis coming from the winners bracket, you are just out.
...Except you would not. You took out extended series but then you forgot how normal double elimination works. Without extended series, Shine would immediately have to rematch against Flash in a brand new best of 3 (And get owned). That's the advantage from making into the finals of a standard double elimination bracket (which is what the semis are in this case).
|
On July 25 2013 22:54 sertman wrote: I don't understand the big fuss about all this anyways. If MLG believes that in order to win their tournament, they must win the majority of games against everybody they play, that's absolutely their perogative. It's OK for two tournaments to have two different philosophies on this matter. If Dreamhack or IEM or ESL or NASL has a different philosophy, that doesn't mean MLG is wrong, just different. It's their tournament, they're spending the money to run it and stream it, they get to say how a player advances.
It's just an excuse for people to hate on MLG more. We all know how much people love that.
It's been a while since I've checked this thread and I found this comment rather odd considering how much praise people gave MLG for hosting an American Invitational and people criticizing Geoff for making a hasty tweet about the thought of SC2 having a smaller stage. Not sure how you missed that, but the responses were overwhelming positive. As for your spiel about tournaments. The CEOs have commented on that matter saying they like having different formats. That's great and all but when it comes down to a global like Blizzard is trying to implement.. I'd make every major offline and implement a World Tour Circuit where you'd get the same format and usual suspects coming to a city near you for unison. I don't mind more games when it comes to Ro8, Ro4 and Finals, but you would have to trim that production time from somewhere. First thing on my list to cut would most likely be double elimination if there was one. That would save a ton of time.
|
On July 23 2013 08:03 KissMeRed wrote: TLDR: It's the same argument from my first post. If you think head to head map scores are important (like I do and I think MLG does) then you are in favor of extended series. If you don't care about map scores and order of series wins (some kind of argument like later rounds are more important) then you probably think extended series is a useless or hurtful rule.
I just want to note that the reason I prefer non extended series, though I'm not extremely opposed apart from in finals, is that to me fairness of all players is more important than fairness between two players. In my opinion, just because a player is lucky/unlucky enough to play against a player that they have already played should not give them an advantage/disadvantage over players who do not meet any players they have already played against. So for me it's not really about the order of series, but the number of full series.
As far as I am concerned, the winner got his advantage by continuing on in the winners bracket and having a better opportunity to win the tournament, once the initial winner loses both players should be on an even field. In the case of a Winner vs Loser, I would favour playing a regular series and a second series if the winner loses, this would mean a minimum of 6 games and a maximum of 9 which is more games than extended series and it also guarantees that whoever advances has at least an even score (2-1, 0-2, 2-1 = 4-4 being the closest possible outcome)
|
On July 25 2013 02:19 Goolpsy wrote:
Look at my example, last post page 4. If the BAD player ends up winning the extended series (knocking out the GOOD player) He will very likely soon meet an opponent of a race he is bad against, giving really really bad games forward.
But if he is really a "Good Player" he should not have problems with opponents of a race he is bad against, unless he is not thet "good".
If he is really good he is going to have problems with his opponents and would just take them like a boss.
|
[note: final math conclusions at bottom of post]
MLG's extended series rule is and always has been a mediocre solution to a very real problem. The issue MLG faces (along with most other major tournaments) is that in a double elimination bracket, a player (or team in the case of dota2, etc) must lose two series to be eliminated. This should be quite obvious from the name. If you look at the recent example of Evo2013, this results in the Grand Final being either a single series won by the Winner's Bracket player, or a two-series final that could be won by either player. Neither player gets a game advantage within a single series, but the player who came from the Winner's Bracket still must lose two full series to drop out of the tournament (as does everyone else).
Let's take some examples from the previous posts and see how they turn out with Evo-style rules. (Disclaimer: I'm no total expert in probability so I'm not 100% sure I handled win probabilities perfectly in bo3/7 format but I think the results will be quite accurate still relative to each other).
Day9 mentions the players GOOD and BAD, where GOOD beats BAD 70% of the time. I'll use P(G) to refer to GOOD winning over and P(~G) to GOOD losing to BAD. All of these probabilities will occur within the sample space of an extended series rematch actually happening.
In a best of 3... P(G) = (3 choose 2)(.7^2)(.3^1) + (3 choose 3)(.7^3)(.3^0) P(G) = (3)(.49)(.3) + (1)(.343) P(G) = .441 + .343 P(G) = .784
Case 1: GOOD wins in WB, is met by BAD in Grand Final. Extended Series: Given GOOD won, GOOD won the WB meetup 2-0 with probability .4375 and 2-1 with probability .5625. P(GOOD wins grand final) = (.4375)(GOOD wins with 2-0 lead) + (.5625)(GOOD wins with 2-1 lead) P(G) = (.4375)((5 choose 2)(.7^2)(.3^3) + (5 choose 3)(.7^3)(.3^2) + (5 choose 4)(.7^4)(.3) + (5 choose 5)(.7^5)) + (.5625)((4 choose 2)(.7^2)(.3^2) + (4 choose 3)(.7^3)(.3) + (4 choose 4)(.7^4)) P(G) = (.4375)(.09261 + .3087 + .36015 + .16807) + (.5625)(.2646+ .4116 + .2401) P(G) = (.4375)(.92953) + (.5625)(.9163) P(G) = .9221 GOOD will win the extended series with probability ~92.21%, which is a good ~14% higher than his chances of winning a regular bo3. It is worth noting that GOOD's chances of winning a simple bo7 vs BAD would actually be 87.40%.
Evo-style rules: GOOD has no game advantage but must lose twice (lose two bo3s) in order to lose the grand final against BAD. P(GOOD wins grand final) = P(GOOD wins first bo3) + P(GOOD loses first bo3)*P(GOOD wins second bo3) P(G) = (.784) + (.216)(.784) P(G) = .95334 GOOD has a 95.33% chance to win the grand finals with Evo-style rules (WB player must lose 2 bo3s)
While GOOD still has a very good chance of winning in either format, Evo-style rules increase GOOD's chances of winning by about 3%. Both formats still give GOOD a nice advantage over both a regular bo3 or bo7.
Case 2: BAD wins in WB, meets GOOD in grand final. Extended Series: Given BAD won in WB, there was a .875 probability it was 2-1 and .125 probability it was 2-0. P(G) = (.875)((4 choose 3)(.7^3)(.3) + (4 choose 4)(.7^4)) + (.125)((5 choose 4)(.7^4)(.3) + (5 choose 5)(.7^5)) P(G) = (.875)(.4116 + .2401) + (.125)(.36015 + .08235) P(G) = .5702375 + .0553125 P(G) = .6256
GOOD will have a 62.56% chance of winning the extended series if he is coming from the loser's bracket. Obviously, this is lower than his chance of winning a bo3 or bo7 as he has already lost two games in this scenario.
Evo-style: GOOD must win two straight bo3s to win the grand final. P(G) = (.784)(.784) P(G) = .6147
GOOD will only have a 61.47% chance of winning two best of threes in a row, which is actually about 1% lower than in an extended series! Maybe we were wrong about you after all, MLG.
Summary & Conclusion
So, we've found out two major things thus far. 1. Extended series gives the better player a slightly better chance to win the full series IF he has already lost the first bo3 to a worse player. 2. Extended series gives the worse player a slightly better chance to win the full series IF he has already lost the first bo3 to a better player. People have been saying that the "worse" player gets a better chance statistically with extended series when the rematch comes around, but the truth is that whoever lost first gets a very slightly better chance for a "comeback" in a potential rematch. Now, let's calculate overall winrates in both cases. Evo-style rules: P(G) = P(GOOD wins first bo3 and grand final) + P(GOOD loses first bo3, wins grand final) P(G) = (.784)(.9533) + (.216)(.6147) P(G) = .8801
Extended Series: P(G) = P(GOOD wins first bo3 and extended series) + P(GOOD loses first bo3, wins extended series) P(G) = (.784)(.9221) + (.216)(.6256) P(G) = .8581
Therefore, GOOD will win his overall series against BAD 88.01% of the time with dual elim Evo-style rules, and 85.81% of the time with Extended Series. It looks like we were right about MLG, and extended series does in fact give GOOD a 2.2% lower chance of winning his series against BAD! This percentage would change if GOOD's bo1 win chance against BAD was different.
Addendum: KissMeRed's scenario is a bit unique in the way it was set up but I will say this in favor of extended series. If Shine was to purposefully lose to Flash, and then Flash was knocked into Lower Bracket to play Shine again (say in LB finals), an extended series would result in Flash winning while a simple bo3 (the regular Evo-style rule) would have him lose to Shine's perfected cheeses. I don't find this scenario particularly realistic (nor do I wish or need to do math with 100% winrates) so I skipped doing any calculations on it.
(edit:typos)
|
The problem with extended series (as I see it?) isn't, the probability for the finals (It is a special case altogether)
But that you insert an inherited disadvange somewhere in the middle of the bracket, with claims of 'fairness' players between, but not taking note, that the competition is between 512 og 1024 players and not just two, and furthermore doesn't take into consideration that the player from the lower bracket has already played a fair deal more opponents than the player from the upper bracket.
|
|
|
|